JP3672371B2 - Method for measuring actual space length by imaging means, optical system calibration method, and reference gauge used for optical system calibration - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an optical system calibration in an image processing system, and in particular, a real space length measurement method, an optical system calibration method, and a reference gauge, which are performed based on an image of a reference gauge imaged by an imaging means via the optical system. It is about.
[0002]
[Prior art]
In recent years, image processing systems have come to be frequently used for measurement and inspection at development sites and manufacturing sites due to progress in speeding up and cost reduction of image processing devices. FIG. 18 is a configuration diagram of a general image processing system.
In the figure, the position of a sample (not shown) placed on the position adjusting means 27 can be adjusted by the means 27. Specifically, the imaging means 24 is a CCD camera, a line sensor, or the like, and images a sample as an image. The optical system 23 forms an image of the sample on the imaging surface of the imaging means. The image memory 25 holds the obtained image. The arithmetic unit 26 performs a predetermined process on the obtained image.
With such a configuration, an image of the sample is taken, the obtained image is processed, and measurement is performed.
[0003]
As an optical system calibration method in such a system, for example, a method is used in which a line pattern or scale having a known length is imaged and the magnification is calculated based on the length of the image. However, due to image distortion caused by the optical system 23 used for imaging, if the obtained image is measured as it is, an error is included in the measurement result. Therefore, it was necessary to correct this distortion. In the technique described in Japanese Patent Laid-Open No. 63-222247, a method for correcting distortion of a captured image in a radiographic apparatus is shown. In this method, a strain measurement member having a radiation absorption coefficient such that the obtained image becomes a dot pattern is imaged. The position of the dot in the obtained image is compared with the position of the true dot that should originally be imaged, and interpolation processing is performed to create a distortion correction table and correct the image.
[0004]
In recent years, audiovisual equipment has been reduced in size as represented by portable audio tape playback apparatuses and movies. On the other hand, video recording equipment such as movies and stationary VTRs are required to record for a long time. An important technique for satisfying such requirements is high-density recording of video and audio signals. In order to realize high-density recording while maintaining compatibility between recording and reproducing devices for each standard such as VTR and DAT, one of the important techniques is to record a track with high linearity. For this reason, a method for inspecting the linearity of the recording track is important. In the technique described in Japanese Patent Application Laid-Open No. 3-222102 as a conventional magnetic recording track inspection apparatus, a recording track is regarded as a lattice, and the bending of the track is considered as a deformation of the lattice, thereby adopting a fringe / lattice image analysis technique. Describes how to measure track linearity. The basic configuration of a conventional magnetic recording track inspection apparatus is almost the same as the configuration example of the image processing system shown in FIG.
[0005]
In such a calibration operation of the conventional magnetic recording track inspection apparatus, the actual space length per pixel is obtained from the set magnification of the optical system 23. In the magnetic recording track inspection apparatus, it is necessary to match the imaging range to the effective area of the track pattern of the magnetic tape (not shown). The adjustment operation of the imaging range is performed by looking at the imaging range of the image taken by the imaging unit 24 and making an approximate adjustment.
Further, the method for confirming the measurement accuracy of the magnetic recording track inspection apparatus obtained as a result of carrying out such a calibration operation was as follows.
First, the displacement distribution in the tape width direction of the track pattern on the magnetic tape is actually measured using a magnetic recording track inspection apparatus. Next, the displacement distribution at the same measurement location on the same magnetic tape is measured by an inspection method using a microscope. In the inspection method using a microscope, the track edge position is measured in the width direction of the magnetic tape by visual observation using a microscope. By comparing this result with the ideal track edge position, the displacement distribution of the track pattern is obtained. The measurement accuracy was confirmed by comparing these two measurement results.
[0006]
[Problems to be solved by the invention]
However, the set magnification of the optical system 23 in FIG. 18 does not exactly match the magnification of the actually obtained image. For this reason, for example, in a measurement where the order of 1 μm or less, such as the measurement of the linearity of the track pattern, becomes a problem, a sufficiently accurate magnification or a real space length per pixel cannot be obtained. Moreover, only the data for one line was obtained by measuring the magnification by imaging the line pattern or scale.
In addition, the conventional imaging range adjustment method can only roughly adjust the imaging range.
Furthermore, with regard to correction of image distortion, in the measurement of image distortion using a dot pattern as described in Japanese Patent Laid-Open No. 63-222247, the number of data acquisition in the measurement area is limited, and the number of interpolation points is increased accordingly. Since it increases, the accuracy deteriorates.
[0007]
Further, the technique described in JP-A-3-222102 does not describe a calibration method. In the measurement accuracy confirmation, a magnetic tape is used as a measurement target. Since the displacement distribution of the track pattern on the magnetic tape is not known, it must be measured using a microscope. However, in the inspection method using a microscope, the measurement accuracy depends on the moving accuracy of the stage on which the magnetic tape is installed. For this reason, the measurement accuracy is at most about ± 0.3 μm, and evaluation with higher accuracy cannot be performed. Also, since magnetic tape is very thin, it will be deformed if an excessive load is applied during handling. For this reason, there is a problem that the displacement distribution changes, and the displacement distribution is not known. For this reason, the accurate measurement accuracy of the magnetic recording track inspection apparatus could not be confirmed.
[0008]
  In view of the above-described problems, the present invention provides a method for precisely measuring the real space length per pixel, and by measuring or adjusting the imaging range, measuring image distortion, and feedback of the result, An object of the present invention is to provide an optical system calibration method for correcting distortion.In the present invention,A magnetic recording track inspection apparatus for realizing inspection of a magnetic recording track that is not affected by image distortion by measuring or adjusting an actual space length per one pixel or an imaging range, measuring image distortion, and feedback of the result. Provides calibration method and reference gaugeDo. Also,Providing high-precision methods for calibrating these optical systems and magnetic recording track inspection equipmentDo.further,Providing more accurate measurement accuracy confirmation method and reference gauge used for measurement accuracy confirmationDo.
[0009]
[Means for Solving the Problems]
The real space length measuring method by the image pickup means according to the present invention is an image pickup means in which an image obtained by forming an image of a grid pattern including a predetermined grid line by an optical system is set to a direction parallel to the grid line and / or a direction perpendicular thereto A grid image is obtained by imaging using an image, and in an arbitrary area in the inspection area of the grid image, the number of grids corresponding to one pixel column of the imaging unit in a direction orthogonal to the grid line is determined for each pixel column. After calculating, the average number of grids is calculated by calculating the average number of grids, and by multiplying the average number of grids by the grid pitch, the real space length of an arbitrary region in the direction orthogonal to the grid lines is calculated. Is divided by the number of pixels in the same direction to calculate the real space length per pixel in the direction orthogonal to the grid line of the grid image. The optical system can be calibrated based on the actual space length per pixel thus obtained.
[0010]
The track pattern of the magnetic recording track on the magnetic tape recorded and visualized by the magnetic recording / reproducing device is imaged on the surface as a reference gauge for the magnetic recording track inspection device for imaging and inspecting by the imaging means. A reference line of a reference gauge coordinate system, and a lattice pattern drawn at an equal pitch so as to have a predetermined angle with respect to the reference direction of the line pattern. A reference gauge is used that has a thickness such that its surface when installed in the inspection device is substantially equal to the height position of its top surface when the magnetic tape is installed. Thereby, confirmation of measurement accuracy can be performed with high accuracy.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
In the real space length measuring method by the imaging means of the present invention, an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch by an optical system in a direction parallel to the lattice lines and And / or a first step of obtaining a lattice image by imaging using an imaging unit with a direction orthogonal to the horizontal scanning direction, and the arbitrary direction in the inspection region of the lattice image in the direction orthogonal to the lattice line The second step of calculating the number of grids corresponding to one pixel column of the image pickup means for each pixel column, and the average number of grids is calculated by averaging the number of grids for each pixel column obtained in the second step. A third step,
A fourth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch; and calculating the real space length of the lattice of the arbitrary region. A fifth step of calculating a real space length per pixel in the direction orthogonal to the grid line of the grid image by dividing by the number of pixels in the direction orthogonal to the line.
[0012]
The second step includes performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area And calculating the number of grids of each pixel column in a direction orthogonal to the line.
[0013]
  Of the present inventionOptical system calibration methodIs a calibration method of an optical system in the case where a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via an optical system, A grid pattern including grid lines parallel to and / or perpendicular to the horizontal scanning direction of the imaging means and arranged at equal intervals with the grid pitch as a value is set at substantially the same position as the imaging position of the magnetic tape. In a first step, a second step of obtaining a lattice image by imaging the lattice pattern using the imaging means, and an arbitrary region in the inspection region of the magnetic recording track inspection device in the lattice image, A third step of calculating, for each pixel column, the number of lattices corresponding to one pixel column of the imaging means in a direction orthogonal to the grid line; and the third step A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column, and multiplying the average number of lattices by the lattice pitch to thereby obtain the arbitrary region in a direction orthogonal to the lattice lines. And calculating the real space length of the grid image by dividing the real space length by the number of pixels in the direction orthogonal to the grid line of the arbitrary region, thereby orthogonal to the grid line of the grid image And a sixth step of calculating a real space length per pixel in the direction.
[0014]
The third step includes performing a Fourier transform in a direction perpendicular to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area And calculating the number of grids of each pixel column in a direction orthogonal to the line.
[0015]
  In the real space length measuring method by the imaging means of the present invention, an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch by an optical system in a direction parallel to the lattice lines and And / or a first step of obtaining a lattice image by imaging using an imaging unit with a direction orthogonal to the horizontal scanning direction, and the arbitrary direction in the inspection region of the lattice image in the direction orthogonal to the lattice line The second step of calculating the number of grids corresponding to one pixel column of the image pickup means for each pixel column, and the average number of grids is calculated by averaging the number of grids for each pixel column obtained in the second step. A third step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch; and A real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated by dividing the inter-space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region. Steps,Real space length per pixelAnd a sixth step of calculating an imaging range in the inspection region by multiplying the number of pixels in a direction orthogonal to the grid line of the inspection region.
[0016]
The second step includes performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area And calculating the number of grids of each pixel column in a direction orthogonal to the line.
[0017]
The sixth step calculates the imaging range in the inspection area of the imaging means by multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the lattice line of the inspection area. The step of adjusting the magnification of the optical system so that the imaging range becomes a predetermined value may be included.
[0018]
The second step includes performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in a direction orthogonal to the line, and the sixth step includes pixels in a direction orthogonal to the grid lines of the inspection area to the real space length per pixel. A step of calculating an imaging range in the inspection area of the imaging means by multiplying by a number, and adjusting a magnification of the optical system so that the imaging range becomes a predetermined value. Kill.
[0019]
  Of the present inventionOptical system calibration methodIs a calibration method of an optical system in the case where a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via an optical system, A grid pattern including grid lines parallel to and / or perpendicular to the horizontal scanning direction of the imaging means and arranged at equal intervals with the grid pitch as a value is set at substantially the same position as the imaging position of the magnetic tape. In a first step, a second step of obtaining a lattice image by imaging the lattice pattern using the imaging means, and an arbitrary region in the inspection region of the magnetic recording track inspection device in the lattice image, A third step of calculating, for each pixel column, the number of lattices corresponding to one pixel column of the imaging means in a direction orthogonal to the grid line; and the third step A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column, and multiplying the average number of lattices by the lattice pitch to thereby obtain the arbitrary region in a direction orthogonal to the lattice lines. And calculating the real space length of the grid image by dividing the real space length by the number of pixels in the direction orthogonal to the grid line of the arbitrary region, thereby orthogonal to the grid line of the grid image A sixth step of calculating a real space length per pixel in the direction, and multiplying the real space length per pixel by the number of pixels in a direction orthogonal to the grid line of the inspection region, And a seventh step of calculating an imaging range in the area.
[0020]
The third step includes performing a Fourier transform in a direction perpendicular to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area And calculating the number of grids of each pixel column in a direction orthogonal to the line.
[0021]
The seventh step calculates the imaging range in the inspection area of the imaging means by multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the lattice line of the inspection area. The step of adjusting the magnification of the optical system so that the imaging range becomes a predetermined value may be included.
[0022]
The third step includes performing a Fourier transform in a direction perpendicular to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in a direction orthogonal to the line, and the seventh step includes pixels in a direction orthogonal to the grid line of the inspection area to the real space length per pixel. A step of calculating an imaging range in the inspection area of the imaging means by multiplying by a number, and adjusting a magnification of the optical system so that the imaging range becomes a predetermined value. Kill.
[0023]
  In the optical system calibration method of the present invention, an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch by means of the optical system is parallel and / or orthogonal to the lattice lines. A first step of obtaining a grid image by imaging using an imaging unit with a horizontal scanning direction as a horizontal scanning direction, and the imaging unit in a direction orthogonal to the grid line in an arbitrary region within the inspection area of the grid image A second step of calculating the number of grids corresponding to one pixel column for each pixel column, and a third step of calculating an average number of grids by averaging the number of grids for each pixel column obtained in the second step. A step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch; and By dividing the number of pixels a direction perpendicular to the grating lines of an arbitrary region, and the fifth step of calculating the grating image, a real space length of 1 per pixel in the direction perpendicular to the grating lines,Real space length per pixelUsing the above, by obtaining the displacement distribution of the grid image in the direction orthogonal to the grid line of the inspection region,Of images resulting from the optical systemA sixth step of obtaining a distortion distribution in the direction over the entire inspection region;
[0024]
  The second step includes performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in the direction orthogonal to the line, and the sixth step includes:The real space length per pixel and the position information of the grid line at each pixelIt is also possible to include a step of calculating a displacement distribution in the direction orthogonal to the lattice lines in the inspection region using the phase value distribution.
[0025]
  The method of calibrating an optical system for imaging a magnetic recording track according to the present invention is a method in which a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via the optical system. A method for calibrating an optical system in a magnetic recording track inspection apparatus for imaging and inspecting, including grid lines arranged at equal intervals with a known value as a grid pitch and parallel and / or perpendicular to the horizontal scanning direction of the imaging means A first step of installing a lattice pattern at substantially the same position as the imaging position of the magnetic tape; a second step of obtaining a lattice image by imaging the lattice pattern using the imaging means; One pixel of the imaging means in a direction orthogonal to the grid line in an arbitrary area within the inspection area of the magnetic recording track inspection apparatus in the lattice image A third step of calculating the number of grids corresponding to each pixel column, and a fourth step of calculating the average number of grids by averaging the number of grids for each pixel column obtained in the third step; A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch, and calculating the real space length of the arbitrary region. A sixth step of calculating a real space length per pixel of the grid image in a direction orthogonal to the grid line by dividing by the number of pixels in a direction orthogonal to the grid line;Real space length per pixelIs used to obtain the distortion distribution in the direction of the optical system of the magnetic recording track inspection apparatus over the entire inspection region by obtaining the displacement distribution in the direction orthogonal to the lattice line in the inspection region of the lattice image. It has a seventh step.
[0026]
  The third step includes performing a Fourier transform in a direction perpendicular to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in the direction orthogonal to the line, and the seventh step includes, The actual space length per pixel, and the position information of the grid line at each pixelThe phase value distribution may be used to calculate a displacement distribution of the lattice image in a direction orthogonal to the lattice line in the inspection region.
[0027]
  In the optical system calibration method of the present invention, an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch by means of the optical system is parallel and / or orthogonal to the lattice lines. A first step of obtaining a grid image by imaging using an imaging unit with a horizontal scanning direction as a horizontal scanning direction, and the imaging unit in a direction orthogonal to the grid line in an arbitrary region within the inspection area of the grid image A second step of calculating the number of grids corresponding to one pixel column for each pixel column, and a third step of calculating an average number of grids by averaging the number of grids for each pixel column obtained in the second step. A step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch; and By dividing the number of pixels a direction perpendicular to the grating lines of an arbitrary region, and the fifth step of calculating the grating image, a real space length of 1 per pixel in the direction perpendicular to the grating lines,Real space length per pixelA sixth step of obtaining a distortion distribution in the direction of the optical system over the entire inspection region by obtaining a displacement distribution in the direction perpendicular to the lattice line of the inspection region of the lattice image using And a seventh step of correcting an image formed by the optical system using the distortion distribution.
[0028]
  The second step includes performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern in the inspection region, extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in a direction orthogonal to the line, and the sixth step includes:The real space length per pixel and the position information of the grid line at each pixelIt is also possible to include a step of calculating a displacement distribution in the direction orthogonal to the lattice lines in the inspection region using the phase value distribution.
[0029]
  The method of calibrating an optical system for imaging a magnetic recording track according to the present invention is a method in which a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via the optical system. A method for calibrating an optical system in a magnetic recording track inspection apparatus for imaging and inspecting, including grid lines arranged at equal intervals with a known value as a grid pitch and parallel and / or perpendicular to the horizontal scanning direction of the imaging means A first step of installing a lattice pattern at substantially the same position as the imaging position of the magnetic tape; a second step of obtaining a lattice image by imaging the lattice pattern using the imaging means; One pixel of the imaging means in a direction orthogonal to the grid line in an arbitrary area within the inspection area of the magnetic recording track inspection apparatus in the lattice image A third step of calculating the number of grids corresponding to each pixel column, and a fourth step of calculating the average number of grids by averaging the number of grids for each pixel column obtained in the third step; A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch, and calculating the real space length of the arbitrary region. A sixth step of calculating a real space length per pixel of the grid image in a direction orthogonal to the grid line by dividing by the number of pixels in a direction orthogonal to the grid line;Real space length per pixelIs used to obtain the distortion distribution in the direction of the optical system of the magnetic recording track inspection apparatus over the entire inspection region by obtaining the displacement distribution in the direction orthogonal to the lattice line in the inspection region of the lattice image. A seventh step and an eighth step of imaging the track pattern recorded on the magnetic tape by the imaging unit and calculating a displacement distribution of the track pattern corrected using the distortion distribution.
[0030]
  The third step includes a step of performing a Fourier transform in a direction orthogonal to a lattice line of the lattice pattern in the inspection region, and extracting a primary frequency component from the obtained frequency spectrum, and extracting the primary frequency component A step of performing inverse Fourier transform and calculating a phase value distribution of the grid image from a ratio of a real part and an imaginary part of the result, and using the phase value distribution, the grid in an arbitrary area in the inspection area Calculating the number of grids of each pixel column in the direction orthogonal to the line, and the seventh step includes:The real space length per pixel and the position information of the grid line at each pixelThe phase value distribution may be used to calculate a displacement distribution of the lattice image in a direction perpendicular to the lattice line in the inspection region.
[0031]
  The lattice pattern in the first step is a reference gauge for the magnetic recording track inspection apparatus, and the reference gauge is
  A line pattern serving as a reference of the reference gauge coordinate system on the surface of the reference gauge;
  A lattice pattern drawn at an equal pitch so as to have a predetermined angle with respect to a reference direction of the line pattern,
  The surface of the line pattern and the lattice pattern when installed in the magnetic recording track inspection apparatus is substantially the same as the height position of the upper surface of the magnetic tape when the magnetic tape is installed in the magnetic recording track inspection apparatus. Have equal thickness.
[0032]
  On the reference gaugeThe pitch of the lattice pattern may be substantially equal to the pitch of the track pattern on the magnetic tape.
[0034]
  Of the present inventionThe calibration method of the magnetic recording track inspection device is as follows:This is an optical system calibration method in a magnetic recording track inspection apparatus in which a track pattern of a magnetic recording track on a magnetic tape recorded and visualized by a magnetic recording / reproducing apparatus is imaged and inspected by an imaging means via an optical system. And
  A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
  A second step of obtaining a lattice image by imaging the lattice pattern using the imaging means;
  In an arbitrary region in the inspection area of the magnetic recording track inspection apparatus in the lattice image, the number of lattices corresponding to one pixel column of the imaging unit in a direction orthogonal to the lattice line is calculated for each pixel column. A third step;
  A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
  A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice lines by multiplying the average number of lattices by the lattice pitch;
  By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, the real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
  By using the real space length per pixel to obtain a displacement distribution in the direction perpendicular to the grid lines in the inspection area of the grid image, the distortion in the direction of the optical system of the magnetic recording track inspection apparatus is obtained. A seventh step of obtaining a distribution over the entire inspection region;
  Of the line pattern serving as a reference of the coordinate system and the ideal track pattern recorded on the magnetic tape by two or more heads each having a predetermined azimuth angle in the magnetic recording / reproducing apparatus, one azimuth angle is determined. A track pattern recorded by a head having a light portion and a track pattern recorded by a head having the other azimuth angle is a dark portion, and is substantially equal to at least one displacement. A pseudo track pattern having a known master displacement distribution with respect to the ideal track pattern along the measurement line, and a surface of the pseudo track pattern when the magnetic tape is installed in the magnetic recording track inspection apparatus. Installed so that it is substantially equal to the height positionAn eighth step to:
  A ninth step of photographing the pseudo track pattern by the magnetic recording track inspection apparatus and calculating a displacement distribution of the pseudo track pattern corrected using the distortion distribution;
  A tenth step of detecting the measurement accuracy of the magnetic recording track inspection apparatus by comparing the displacement distribution and the master displacement distribution;
Have
[0035]
  In calculating the displacement distribution in the ninth step,
  In the image of the pseudo track pattern, Fourier transform is performed in the displacement distribution measurement direction, and a primary frequency component is extracted from the obtained frequency spectrum,
  Inverse Fourier transform is performed on the extracted primary frequency component, and the phase value distribution of the pseudo track pattern image is calculated from the ratio of the real part and the imaginary part of the result,
  The displacement distribution of the pseudo track pattern image can be calculated using the phase value distribution.
[0036]
  The lattice pattern in the first step is a reference gauge for the magnetic recording track inspection device,
  The reference gauge is a line pattern serving as a reference of the coordinate system of the reference gauge on the surface of the reference gauge;
  A lattice pattern drawn at an equal pitch so as to have a predetermined angle with respect to a reference direction of the line pattern,
  In the magnetic recording / reproducing apparatus, among the ideal track patterns recorded on the magnetic tape by two or more heads each having a predetermined azimuth angle, the track pattern recorded by the head having one azimuth angle is clarified. And a pseudo track pattern substantially equal to the track pattern obtained when the track pattern recorded by the head having the other azimuth angle is a dark portion,
  A displacement distribution of the pseudo track pattern with respect to the ideal track pattern along at least one or more displacement measurement lines on the pseudo track pattern is known;
  The surface of the line pattern, the lattice pattern, and the pseudo track pattern when the magnetic tape is installed in the magnetic recording track inspection apparatus is the upper surface of the magnetic tape when the magnetic tape is installed in the magnetic recording track inspection apparatus. Substantially equal to the height position ofUsing the reference gauge having a thickness of.
[0037]
【Example】
Example 1
Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is a block diagram of an optical system calibration system. In the figure, a predetermined lattice pattern 7 to be described later is drawn on the surface of the reference gauge 1. The CCD camera 3 as an imaging unit is arranged so as to share an optical axis with the optical system 2 composed of a single lens or a lens group, and captures an image formed by the optical system 2. The image memory 4 stores the image from the CCD camera 3, and the arithmetic unit 5 performs processing on the image stored in the image memory 4. The position adjusting means 6 of the reference gauge 1 is, for example, an XYθ stage.
FIG. 2 is a diagram showing details of the lattice pattern 7 described above. As shown in FIG. 2, a plurality of lattice lines 7a are formed so that the lattice pitch is equal, and the intervals are known values.
[0038]
The procedure of the optical system calibration method in the optical system calibration system configured as described above will be described with reference to the functional block diagram of FIG.
First, imaging of a lattice image (step 101) will be described. First, the position of the reference gauge 1 is adjusted using the XYθ stage 6 so that the lattice line 7a (FIG. 2) of the reference gauge 1 shown in FIG. 1 is parallel to the horizontal scanning direction of the CCD camera 3. Next, the reference gauge 1 is imaged by the CCD camera 3. FIG. 4 is a diagram showing a lattice image 8. The lattice image 8 is obtained by imaging the lattice pattern 7 on the reference gauge 1 with the CCD camera 3 and is stored in the image memory 4. As shown in FIG. 4, in the lattice image 8, the direction corresponding to the horizontal scanning direction of the CCD camera 3 in FIG. 1 is defined as the X direction, and the direction corresponding to the vertical scanning direction is defined as the Y direction. In the calculation of the actual space length per pixel, the imaging range, and the distortion distribution, the number of grids in the inspection region 9 is related to the measurement resolution, and the accuracy is improved with a larger number of grids. Therefore, when imaging with the reference gauge 1, it is desirable that each grid line captures a large number of grids within the resolution of the CCD camera 3. For example, when the number of pixels in the direction orthogonal to the grid lines of the grid image 8 (the Y direction in FIG. 4) is 512, the number of grids is preferably about 128.
[0039]
Next, calculation of the number of grids (step 102) in FIG. 3 will be described. First, an area to be calibrated, that is, an inspection area 9 is determined in the lattice image 8 of FIG. The actual space length, the imaging range, the distortion distribution, and the like per pixel in the inspection area are to be calibrated. Accordingly, the inspection area 9 may be set to an area mainly used in the image formed by the optical system 2 in FIG. For example, in the case of an optical system incorporated in a measurement apparatus, an area used for imaging the measurement target may be set. An arbitrary area 10 is set in the inspection area 9. In this region, the actual space length per pixel is calculated. The calculation accuracy of the real space length per pixel is improved when the number of grids is large. Therefore, it is desirable that the number of grids included in the arbitrary region 10 is as many as possible within the resolution of the CCD camera 3 for each grid line. The inspection area 9 and the arbitrary area 10 may be the same.
[0040]
In the arbitrary region 10 in the inspection region 9 in the lattice image 8 stored in the image memory 4, the number of lattices included for each pixel column in the Y direction is calculated by the arithmetic device 5. The arithmetic device 5 calculates the number of grids by a method as described below, for example. First, binarization processing is performed on the image in the arbitrary area 10. The binarization process is performed as follows. First, the threshold value of the luminance distribution in the arbitrary area 10 is determined. The threshold value may be an average value of the luminance distribution in the arbitrary region 10. Pixels that are higher than the threshold value are bright and pixels that are lower than the threshold value are dark. With the above processing, the image of the arbitrary area 10 is converted into a binary image of a bright part and a dark part. An image obtained by imaging the reference gauge 1 generally has a good contrast and little noise. Therefore, when binarization processing is performed, the luminance distribution becomes a complete rectangular wave. The number of lattices is calculated by counting the number of waves of this rectangular wave.
When the number of lattices is calculated in units of integers in this way, it is desirable to set the arbitrary region 10 so that the number of lattices included in the arbitrary region 10 is as integer as possible.
[0041]
Next, the calculation of the average number of grids (step 103) in FIG. 3 will be described. In the arithmetic unit 5 of FIG. 1, the average number of lattices is calculated by calculating the average value of the number of lattices in each obtained pixel column. Although the number of grids varies depending on the location due to the influence of image distortion, the average grid number calculation in step 103 calculates the average grid number in the two-dimensional area 10 of FIG. The method of performing such a measurement calculates the actual space length per pixel in FIG. 3 (step 105) and the imaging range rather than performing the measurement with one line as in the method of obtaining the magnification using the line pattern or scale. With respect to the actual space length per one pixel and the imaging range calculated in the calculation (step 106), a representative value that more appropriately represents the lattice image 8 can be obtained.
[0042]
Next, the real space length calculation (step 104) in FIG. 3 will be described. In the arithmetic unit 5 in FIG. 1, the real space length in the Y direction of the arbitrary region 10 in FIG. 4 is calculated by multiplying the average number of lattices by the known pitch length of the lattice pattern 7 on the reference gauge 1.
Next, the calculation of the actual space length per pixel (step 105) in FIG. 3 will be described. In the arithmetic device 5 of FIG. 1, the real space length per pixel is obtained by dividing the real space length in the Y direction of the arbitrary region 10 in FIG. calculate.
[0043]
Next, the imaging range calculation (step 106) in FIG. 3 will be described. In the arithmetic unit 5 in FIG. 1, the real space length per one pixel is multiplied by the number of pixels in the Y direction of the inspection region 9 in FIG. A range can be calculated.
In addition, when the magnification of the optical system 2 in FIG. 1 can be continuously changed as in a stereomicroscope, the magnification adjustment operation can be performed based on the calculated imaging range. When the imaging range to be set is determined in advance, the magnification is calculated from the actual space length of the imaging range and the actual space length of the image sensor of the CCD camera 3. After setting the optical system 2 to the magnification, the imaging range is calculated. In general, the imaging range of the calculation result does not match the desired imaging range due to the influence of image distortion or the like. The imaging range is adjusted by repeatedly adjusting the magnification of the optical system 2 and calculating the imaging range so that they match.
[0044]
Next, the distortion distribution calculation (step 107) in FIG. 3 will be described. In the arithmetic unit 5 of FIG. 1, using the real space length per pixel calculated by the calculation of the real space length per pixel (step 105), the displacement distribution in the inspection region 9 of the lattice image 8 of FIG. The distortion distribution of the image caused by the optical system 2 is calculated. As a method for calculating the displacement distribution, for example, binarization processing is performed on the image of the inspection region 9. Further, a thinning process as shown in FIG. 5 is performed, and the center position of the grid line is obtained for each pixel. By multiplying the obtained position distribution by the actual space length per pixel, the position distribution of the lattice lines in units of the actual space length is obtained. The displacement distribution in the Y direction of the lattice image 8 shown in FIG. 4 is calculated by comparing the obtained position distribution with the true position distribution calculated from the pitch of the lattice pattern 7 in FIG. The displacement data obtained by the above processing is only the center position of the grid line in the inspection region 9. Interpolation processing such as spline interpolation is performed to obtain a displacement distribution between grid lines.
[0045]
In such pixel-by-pixel measurement, the measurement accuracy of the displacement distribution depends on the lattice pitch of the reference gauge 1 and the actual space length per pixel. The narrower the grid pitch, the better the resolution of the displacement, and the shorter the real space length per pixel, the better the resolution of the grid line position. On the other hand, the grating pitch can be narrowed only to a level at which each grating line can be identified by the optical system or to the limit of the accuracy of creating the reference gauge 1. Further, it is necessary to secure the number of pixels of the CCD camera 3 accordingly. Considering these, it is necessary to determine the lattice pitch of the reference gauge 1 and the number of pixels of the image pickup means such as the CCD camera 3. For example, assume that the resolution of the optical system 2 to be used is 8 μm. In this case, the lattice pitch of the reference gauge 1 is required to be at least 16 μm. For example, if the grid pitch is 32 μm and the number of pixels of the CCD camera 3 to be used is 512 pixels in the horizontal and vertical directions, the number of grids to be imaged is preferably about 128, so the imaging range is about 4 mm square. It becomes an area.
[0046]
The above description is all about the processing in the Y direction of the inspection region 9 in FIG. After adjusting the position of the reference gauge 1 using the XYθ stage 6 so that the grid line of the reference gauge 1 in FIG. 1 is parallel to the vertical scanning direction of the CCD camera 3, and imaging the reference gauge 1 with the CCD camera 3, By repeating the same procedure as described above, the real space length, the imaging range, and the distortion distribution per pixel in the X direction can be calculated. Note that lattice patterns orthogonal to each other may be drawn on the reference gauge 1 in advance as shown in FIG. Alternatively, the lattice pattern 7 may be a lattice in two directions as shown in FIG. 7, and processing in the X and Y directions of a lattice image obtained by imaging the lattice pattern 7 may be performed simultaneously.
[0047]
Next, image capturing (step 108) in FIG. 3 will be described. In the imaging of the image, an image of the measurement object is captured using the optical system 2 and the CCD camera 3 of FIG.
Next, the image correction (step 109) in FIG. 3 will be described. In the arithmetic unit 5 of FIG. 1, the correction of the image using the obtained X, Y direction distortion distribution is performed. The image to be corrected is an arbitrary image captured in image capturing (step 108).
FIG. 8 is a schematic diagram for explaining a method of correcting the luminance value of a certain pixel point. P (0,0), P (1,0), P (0,1) and P (1,1) are each pixel. A case where the luminance of the pixel P (0,0) is to be corrected will be described. The position vector of each pixel point is set to P (0,0), P (1,0), P (0,1), and P (1,1) using the symbols as they are. D (0,0), D (1,0), D (0,1), and D (1,1) are pixels P (0,0), P (1,0), P (0, It is a vector representing the amount of distortion at 1) and P (1,1). At this time, the luminances at P (0,0), P (1,0), P (0,1) and P (1,1) are respectively B (0,0), B (1,0), B Let (0,1) and B (1,1), and the corrected luminance at P (0,0) be Bt (0,0). The corresponding positions of each pixel P (0,0), P (1,0), P (0,1) and P (1,1) after distortion correction are respectively Pt (0,0) and Pt (1,0 ), Pt (0,1) and Pt (1,1). The respective position vectors Pt (0,0), Pt (1,0), Pt (0,1) and Pt (1,1) are obtained by the following equations.
[0048]
Pt (0,0) = P (0,0) −D (0,0)
Pt (1,0) = P (1,0) -D (1,0)
Pt (0,1) = P (0,1) −D (0,1)
Pt (1,1) = P (1,1) −D (1,1)
Here, Pt (0,0), Pt (1,0), Pt (0,1) and the distance from Pt (1,1) to P (0,0) are d (0,0) and d, respectively. Assuming (1,0), d (0,1), and d (1,1), the result is as follows.
d (0,0) = | P (0,0) −Pt (0,0) |
d (1,0) = | P (0,0) −Pt (1,0) |
d (0,1) = | P (0,0) −Pt (0,1) |
d (1,1) = | P (0,0) −Pt (1,1) |
[0049]
Constants K, L, M, and N are defined by the following equations.
K = (1 / d (0,0)) / (1 / d (0,0) + 1 / d (1,0) + 1 / d (0,1) + 1 / d (1,1))
L = (1 / d (1,0)) / (1 / d (0,0) + 1 / d (1,0) + 1 / d (0,1) + 1 / d (1,1))
M = (1 / d (0,1)) / (1 / d (0,0) + 1 / d (1,0) + 1 / d (0,1) + 1 / d (1,1))
N = (1 / d (1,1)) / (1 / d (0,0) + 1 / d (1,0) + 1 / d (0,1) + 1 / d (1,1))
At this time, Bt (0,0) is obtained by the following equation.
Bt (0,0) = K · B (0, 0) + L · B (1,0) + M · B (0, 1) + N · B (1, 1)
As described above, in each pixel in the inspection area 9, the luminance is calculated using four points in the vicinity of the target pixel among the corrected pixels calculated using the distortion distribution, thereby correcting the image. Can be implemented.
[0050]
Thus, according to the first embodiment, the actual space length per pixel when an image formed by the optical system 2 is picked up by the CCD camera 3 can be accurately calculated. By multiplying the actual space length per pixel by the number of pixels in the direction orthogonal to the lattice lines of the inspection area 9, the image formed by the optical system 2 is captured in the inspection area 9 when captured by the CCD camera 3. The imaging range of the corresponding part can be calculated.
In addition, the imaging range in the inspection area 9 of the CCD camera 3 is calculated by multiplying the actual space length per pixel by the number of pixels in the direction orthogonal to the grid lines of the inspection area 9, and the imaging range becomes a predetermined value. By adjusting the magnification of the optical system 2 in such a manner, the imaging range of a portion corresponding to the inspection area 9 when an image formed by the optical system 2 is captured by the CCD camera 3 can be adjusted.
[0051]
Further, the displacement distribution in the direction orthogonal to the lattice line of the inspection region 9 is obtained using the actual space length per pixel, and the distortion distribution in the direction orthogonal to the lattice line of the optical system 2 is obtained over the entire inspection region 9. Thus, the distortion distribution of the portion corresponding to the inspection area 9 when the image formed by the optical system 2 is picked up by the CCD camera 3 can be calculated.
Further, by correcting the image formed by the optical system 2 using the distortion distribution, the image when the image formed by the optical system 2 is picked up by the CCD camera 3 can be corrected.
[0052]
Example 2
Next, a second embodiment will be described. In this embodiment, phase analysis using Fourier transform is used for calibration of the optical system.
The apparatus used in the second embodiment is the same as the optical system used in the first embodiment. The procedure of the optical system calibration method is also the same as the procedure of the optical system calibration method of the first embodiment. However, in the second embodiment, phase information processing using Fourier transform is used in calculating the number of lattices (step 102) in FIG. This will be described.
First, in the arbitrary region 10 in the inspection region 9 in the lattice image 8 (FIG. 4) stored in the image memory 4 of FIG. 1, the number of lattices included for each pixel column in the Y direction is calculated by the arithmetic unit 5. The arithmetic device 5 uses a method of phase information processing of the lattice image 8 using Fourier transform described below.
[0053]
FIG. 13 is a graph showing an example of a waveform of a luminance distribution of one line in a direction orthogonal to the grid line in the X direction or the Y direction in the inspection region 9 of the grid image 8. On the other hand, a frequency spectrum is obtained by performing Fourier transform.
FIG. 14 shows a schematic diagram of a power spectrum that is a sum of squares of a real part and an imaginary part of a frequency spectrum obtained when the luminance distribution of FIG. 13 is subjected to Fourier transform. From this frequency spectrum, only the primary frequency component (corresponding to the shaded portion in FIG. 14) representing the primary harmonic component of the original waveform is extracted, and when the inverse Fourier transform is performed, the original waveform is a smooth waveform in the real part. And the imaginary part is obtained by shifting the waveform of the real part by half a wavelength. If the arctangent of the imaginary part divided by the real part is taken, the phase value of the first harmonic wave of the original waveform at each pixel is obtained. Of the phase value distribution in the inspection region 9 of the lattice image 8 obtained in this way, the amount of change in the phase value in the arbitrary region 10 is calculated. Since the phase value change amount is 2π, which corresponds to one lattice, if the phase value change amount is divided by 2π, the number of lattices included in the arbitrary region 10 can be calculated with a precision below the decimal point. Therefore, it is possible to obtain a value with higher accuracy than a method using binarization processing.
[0054]
Next, phase information processing using Fourier transform is used in the distortion distribution calculation (step 107) of FIG. This will be described.
In the arithmetic unit 5, the displacement distribution in the inspection region 9 of the lattice image 8, that is, the image caused by the optical system is used by using the actual space length per pixel obtained in the calculation of the actual space length per pixel (step 105). Calculate the distortion distribution. If the phase calculation in the inspection region 9 of the lattice image 8 has already been performed at the time of calculating the number of lattices (step 102), the calculation result is used. If phase calculation is not performed, this is carried out by the procedure described in the case of using phase information processing using Fourier transform in calculation of the number of lattices (step 102), and the phase value distribution in the inspection region 9 of the lattice image 8 Is calculated in advance. The lattice image 8 is obtained by taking the difference between the position distribution of each pixel calculated from the actual space length per pixel and the obtained phase value distribution divided by 2π for each pixel and multiplying by the lattice pitch length. The displacement distribution in the direction perpendicular to the grid lines is calculated. In the arithmetic processing using the Fourier transform described above, the displacement data is obtained at all the pixel points in the inspection region 9 and no interpolation processing is required. Further, since the position distribution can be calculated with an accuracy of a pixel unit or less, the distortion can be calculated with high accuracy even when the distortion is 1 pixel or less. For example, even when an image is captured with a real space length of 5 μm per pixel, a distortion amount of 5 μm or less can be calculated.
[0055]
As described above, according to the second embodiment, since the number of lattices is obtained with high accuracy up to a decimal unit by using phase information processing using Fourier transform when calculating the number of lattices, the real space length per pixel is measured. And imaging range calculation and adjustment can be performed with higher accuracy.
In addition, since the displacement distribution can be calculated in decimal pixel units by using phase information processing using Fourier transform when calculating the distortion distribution, the distortion distribution calculation and the image correction using the distortion distribution can be performed with higher accuracy. Can do.
[0056]
Example 3
Hereinafter, a third embodiment will be described with reference to the drawings.
FIG. 9 is a block diagram showing the magnetic recording track inspection apparatus 20. Predetermined patterns 41 to 43 are drawn on the surface of the reference gauge 21. The predetermined patterns 41 to 43 are, as shown in FIG. 10, a line pattern 41 serving as a reference of the pattern, and the lattice pitch is equally spaced and known. Desirably, the lattice pitch is a track of the same azimuth on the magnetic tape. A grid pattern 42 parallel to the line pattern 41 and a grid pattern 43 orthogonal to the line pattern 41, which is equal to the pitch (for example, 20 μm in the case of the DVC format).
[0057]
A reference gauge 21 and a magnetic tape (not shown) are installed on the sample mounting table 22. The reference gauge 21 has a thickness and flatness so that the height of the reference gauge 21 is the same as that when the magnetic tape is set when the reference gauge 21 is set on the sample setting table 22. As a result, the optical system of the magnetic recording track inspection apparatus 20 can be calibrated using the reference gauge 21. The CCD camera 3 as an imaging unit is arranged so as to share an optical axis with the optical system 2 composed of a single lens or a lens group, and captures an image formed by the optical system 2. An image from the CCD camera 3 is stored in the image memory 4, and the arithmetic unit 5 processes the image stored in the image memory 4. The position of the reference gauge 21 is adjusted by the position adjusting means 6. This position adjusting means 6 is, for example, an XYθ stage.
[0058]
The procedure of the optical system calibration method in the magnetic recording track inspection apparatus configured as described above will be described with reference to the functional block diagram of FIG.
First, the installation (step 201) of the reference gauge in FIG. 11 will be described. That is, the reference gauge 21 of FIG. In order to make the height and flatness of the reference gauge 21 and the magnetic tape (not shown) equal, the reference gauge 21 may be made with the same thickness and flatness as the magnetic tape. However, in actuality, the thickness of the magnetic tape is very thin, from several μm to several tens of μm, and it is difficult to create the reference gauge 21 with that thickness. Therefore, the reference gauge 21 is made thicker by a certain value than the thickness of the magnetic tape. In order to absorb the difference in height with the sample mounting table 22, a plate having a thickness difference between the height of the magnetic tape and the reference gauge 21 is incorporated in the sample mounting table 22, and the plate is attached to the magnetic tape when the magnetic tape is installed. To lay under. Alternatively, an elevating mechanism (not shown) in the optical axis direction of the optical system 2 may be provided on the sample setting table 22.
[0059]
Next, imaging of the lattice image (step 202) in FIG. 11 will be described. First, after the position of the reference gauge 21 is adjusted using the XYθ stage 6 so that the line pattern 41 (FIG. 10) of the reference gauge 21 in FIG. 9 is parallel to the horizontal scanning direction of the CCD camera 3, the reference gauge 21 is adjusted. Each of the lattice patterns 42 and 43 on the image 21 is imaged by the CCD camera 3. For example, in the case of the lattice pattern 42, the image captured by the CCD camera 3 and stored in the image memory 4 is as a lattice image 8 shown in FIG. At this time, the reference gauge 21 is installed at the same position as the magnetic tape. The lattice pitch of the reference gauge 21 is the same as the track pitch of the magnetic tape. Thus, the luminance distribution of the obtained lattice image 8 is close to the luminance distribution when the magnetic tape is imaged. Therefore, calibration can be executed under conditions close to actual measurement, which is desirable as a calibration method.
[0060]
Hereinafter, processing of the lattice image 8 obtained when the lattice pattern 42 is imaged will be described. Therefore, a case where processing such as calculation of the number of grids is performed in the Y direction will be described. In the case of the lattice pattern 43, the processing direction may be read as the X direction.
Next, the calculation of the number of grids (step 203) in FIG. 11 will be described. In the grid image 8 of FIG. 4, an area to be calibrated, that is, an inspection area 9 is determined. The actual space length, imaging range, distortion distribution, and the like of this inspection area are to be calibrated. In the magnetic recording track inspection apparatus 20, the inspection area 9 is set so as to substantially coincide with the effective area of the magnetic tape to be measured, that is, the area where the track pattern is drawn. For example, in the case of the DVC format, it is 5.24 mm. If the pitch of the grating is 20 μm, the number of gratings included in the inspection region 9 is 262. Therefore, the number of pixels in the inspection area 9 is desirably 1000 pixels or more. Next, an arbitrary area 10 is set in the inspection area 9. In this region, the actual space length per pixel is calculated. The inspection area 9 and the arbitrary area 10 may be the same.
In the arbitrary region 10 in the inspection region 9 in the lattice image 8 stored in the image memory 4, the number of lattices included for each pixel column in the Y direction is calculated by the arithmetic unit 5 in FIG. The arithmetic device 5 calculates the number of grids by the method described in the calculation of the number of grids (step 102) of the first embodiment.
[0061]
Next, calculation of the average number of grids (step 204) in FIG. 11 will be described. Similar to the processing in the average grid number calculation (step 103) of the first embodiment, the calculation unit 5 in FIG. 9 calculates the average grid number by calculating the average value of the grid numbers in each obtained pixel column. To do.
Next, the real space length calculation (step 205) in FIG. 11 will be described. Similar to the processing in the real space length calculation (step 104) of the first embodiment, the arithmetic unit 5 in FIG. 9 multiplies the average number of grids by the known pitch length of the grid pattern 42 on the reference gauge 21. The real space length in the Y direction of the arbitrary area 10 is calculated.
[0062]
Next, the calculation of the actual space length per pixel (step 206) in FIG. 11 will be described. Similar to the processing in the calculation of the real space length per pixel (step 105) in the first embodiment, the arithmetic device 5 in FIG. The actual space length per pixel is calculated by dividing by the number of pixels in the Y direction.
Next, the imaging range calculation (step 207) in FIG. 11 will be described. Similar to the processing in the imaging range calculation 106 of the first embodiment, the Y direction of the inspection area 9 is set to the real space length per pixel obtained by the real space length calculation 206 per pixel in the arithmetic device 5 of FIG. The actual space length in the Y direction of the inspection area 9, that is, the imaging range of the inspection area can be calculated.
Similarly to the case of the first embodiment, it is also possible to adjust the magnification of the optical system 2 having a variable magnification so that the imaging range matches the effective area of the magnetic tape based on the calculated imaging range.
[0063]
Next, the distortion distribution calculation (step 208) in FIG. 11 will be described. Similar to the processing in the distortion distribution calculation (step 107) of the first embodiment, the real space length per pixel obtained in the real space length calculation per pixel (step 206) is used in the arithmetic unit 5 of FIG. Thus, the displacement distribution in the inspection region 9 of the lattice image 8 in FIG. 4, that is, the distortion distribution of the image caused by the optical system 2 is calculated.
It should be noted that instead of the grid patterns 42 and 43 (FIG. 10) on the reference gauge 21, a grid in the two directions as shown in FIG. May be.
[0064]
Next, imaging of the track pattern image (step 209) in FIG. 11 will be described. First, the magnetic tape is set on the sample setting table 22 of FIG. At this time, the magnetic tape is in the same position as the reference gauge 21. The magnetic tape is previously visualized. Position adjustment is performed using the XYθ stage 6 so that the edge of the magnetic tape is parallel to the horizontal scanning direction of the CCD camera 3. Tracks recorded by two types of magnetic heads having different azimuth angles, that is, head gap angles, are recorded on the magnetic tape so as to be alternately arranged. Illumination is performed so that the track having one azimuth angle is a bright portion and the other track having an azimuth angle is a dark portion. For this reason, it becomes a light and dark pattern just like a lattice pattern. This pattern is imaged by the CCD camera 3. Since the tracks are recorded with a slight inclination from the longitudinal direction of the magnetic tape, the resulting track pattern image 28 is as shown in FIG. As shown in FIG. 12, the longitudinal direction of the tape is the X direction and the tape width direction is the Y direction. The track pattern image 28 is stored in the image memory 4.
[0065]
Next, calculation of the displacement distribution of the track pattern excluding the influence of image distortion in FIG. 11 (step 210) will be described. First, the displacement distribution in the Y direction of the track pattern is calculated according to the displacement distribution calculation of the magnetic recording track inspection apparatus 20 using the track pattern image 28 obtained in the imaging of the track pattern image (step 209). The displacement distribution obtained at this time still includes the influence of image distortion.
FIG. 15 is an explanatory diagram of a method of correcting the displacement distribution of the track pattern. Vector S_EXAnd S_EYIndicates the X and Y components of the distortion of the image at a pixel P in the track pattern image 28, respectively. These are all displacements in the Euler coordinate system and represent where the pixel P has been displaced. Also vector S_LX, S_LYRepresents the displacement of the image distortion in the pixel P in the Lagrangian coordinate system, that is, where the point that should exist in the pixel P has been displaced. Here, assuming that the differential value of the distortion distribution is sufficiently small, S_LXAnd S_LYEach S_EXAnd S_EYIs approximated by
[0066]
By doing so, the influence of the Y direction component of the distortion of the image included in the vector D (not shown) representing the displacement in the Y direction of the track pattern in the pixel P is S_EYIt becomes itself. The influence of the X direction component of the distortion of the image included in the vector D is based on the assumption that the differential value of the distortion distribution is sufficiently small._EXYIs approximated by In FIG. 15, θ is the inclination angle of the track. Therefore, the vector Dt representing the displacement in the Y direction of the track pattern after correction is
Dt = DS_EY+ S_EX× Tanθ
As obtained. The calculation device 5 calculates the displacement distribution of the track pattern excluding the influence of the distortion of the image by performing this calculation for all the pixels of the displacement distribution including the influence of the distortion of the image.
[0067]
As another method of calculating the displacement distribution of the track pattern (step 210) excluding the influence of image distortion, the following processing may be performed by the arithmetic unit 5. That is, the track pattern image 28 obtained in the imaging of the track pattern image (step 209) is corrected using the distortion distribution in the X and Y directions by the method described in the image correction (step 109) of the first embodiment. I do. The obtained track pattern image 28 after correction is subjected to binarization processing and thinning processing, and the center position of the grid line is obtained in units of pixels. By multiplying the obtained position distribution by the actual space length per pixel, the position distribution of the track pattern in units of the actual space length is obtained. The displacement distribution of the track pattern image 28 is calculated by comparing the obtained position distribution of the track pattern and the position distribution of the ideal track pattern. Interpolation processing such as spline interpolation is performed to obtain a displacement distribution between grid lines.
[0068]
Next, a description will be given of a method for confirming the measurement accuracy of the magnetic recording track inspection apparatus obtained as a result of calibrations such as calculation of actual space length per pixel, adjustment of imaging range, and calculation of distortion distribution.
First, in the same manner as the installation of the reference gauge of FIG. 11 (step 201), a measurement accuracy confirmation reference gauge 44 as shown in FIG. 16 is installed on the sample installation table 22 of the magnetic recording track inspection apparatus 20. As in the case of the reference gauge 21, the measurement accuracy confirmation reference gauge 44 is also formed to have the same thickness as the magnetic tape, or thicker than the magnetic tape by a certain value. For this reason, the position of the upper surface of the magnetic tape when the magnetic tape is installed in the magnetic recording track inspection apparatus 20 is the same as the position of the pattern drawing surface of the reference gauge 44 for measuring accuracy confirmation.
[0069]
A pattern as shown in FIG. 16 is drawn on the reference gauge 44 for measuring accuracy. This pattern includes a line pattern 41 and a pseudo track pattern 45. The pitch of the pseudo track pattern 45 is drawn to be equal to the ideal pitch of the same azimuth track of the track pattern recorded on the magnetic tape by the magnetic recording / reproducing apparatus. The angle (θ in FIG. 16) of the pseudo track pattern 45 with respect to the line pattern 41 is selected to be equal to the ideal track angle of the track pattern recorded on the magnetic tape by the magnetic recording / reproducing apparatus. For example, in the DVC format, the pitch is 20 μm and the track angle is 9.166809 °. The line pattern 41 is used as a reference when adjusting the position when the pseudo track pattern 45 is imaged.
[0070]
Next, the magnetic recording track inspection device 20 captures an image of the measurement accuracy confirmation reference gauge 44 and calculates the displacement distribution of the pseudo track pattern 45 excluding the influence of the distortion of the image. When the measurement accuracy confirmation reference gauge 44 is imaged, a pseudo track pattern image (not shown) similar to the track pattern image 28 is obtained. The pseudo track pattern 45 in which the influence of the distortion of the image is eliminated by performing the same processing as the calculation of the displacement distribution of the track pattern in which the influence of the distortion of the image in FIG. The displacement distribution (displacement distribution in the Y direction in FIG. 16) with respect to the ideal track pattern is calculated.
The displacement distribution of the pseudo track pattern 45 may be calculated using phase information processing using Fourier transform described in the description of the distortion distribution calculation (step 107) of the second embodiment.
[0071]
Further, the displacement distribution of the pseudo track pattern 45 with respect to the ideal track pattern is measured by other measuring means. Data obtained by this measuring means becomes master data for accuracy confirmation. Therefore, the measurement means must have a measurement accuracy better than that of the magnetic recording track inspection apparatus. The reference gauge 44 for checking measurement accuracy is created by the same process as that for creating a mask used in the semiconductor field. At this time, the dark part of the pseudo track pattern 45 is made of chromium formed on the glass substrate. Further, nothing is processed in the bright portion of the pseudo track pattern 45. Therefore, the glass substrate can be seen. There is a light wave interference type coordinate measuring machine as a measuring machine for measuring the position of a line drawn with chrome on a glass substrate with high accuracy. According to this, the edge position of the mask pattern can be measured with a repeatability of 0.01 μm or less. Using this measuring machine, the position distribution of the lattice lines of the pseudo track pattern 45 of the reference gauge 44 for measuring accuracy confirmation is measured, and the displacement distribution with respect to the ideal track pattern is calculated. Hereinafter, the displacement distribution obtained in this way is referred to as a master displacement distribution. The master displacement distribution has sufficient accuracy to evaluate the measurement accuracy of the magnetic recording track inspection apparatus in sub μm. Moreover, the displacement distribution of the pseudo track pattern 45 can be maintained by creating the measurement accuracy confirmation reference gauge 44 on the glass substrate. Therefore, by obtaining the master displacement distribution as described above, the displacement distribution of the pseudo track pattern 45 becomes known.
[0072]
Next, from the displacement distribution obtained by the magnetic recording track inspection apparatus, data corresponding to the master displacement distribution, that is, data on the same measurement line is extracted. This is called a sample displacement distribution. By comparing the sample displacement distribution and the master displacement distribution, it is possible to confirm the measurement accuracy of the magnetic recording track inspection apparatus with sub-μm or less accuracy.
As shown in FIG. 17, the line pattern 41, the Y-direction lattice pattern 42, the X-direction lattice pattern 43, and the pseudo track pattern 45 may be drawn on one reference gauge. By using such a reference gauge, real space length calculation per pixel, calculation and adjustment of imaging range, distortion distribution calculation, and measurement accuracy confirmation can be performed without replacing the reference gauge.
[0073]
Thus, according to the third embodiment, the actual space length per pixel of the image obtained by the magnetic recording track inspection apparatus 20 can be calculated.
Further, the imaging range in the inspection area 9 is calculated by multiplying the actual space length per pixel by the number of pixels in the direction orthogonal to the lattice line of the inspection area 9, and is obtained by the magnetic recording track inspection apparatus 20. The imaging range of the part corresponding to the inspection area 9 of the image can be calculated.
Also, by multiplying the actual space length per pixel by the number of pixels in the direction orthogonal to the grid lines of the inspection area 9, the imaging range in the inspection area 9 of the CCD camera 3 is calculated, and the imaging range is a predetermined value. By adjusting the magnification of the optical system 2 so as to be, the imaging range of the portion corresponding to the inspection area 9 of the image obtained by the magnetic recording track inspection apparatus 20 can be adjusted.
[0074]
Also, using the actual space length per pixel, a displacement distribution in the direction orthogonal to the lattice line in the inspection region 9 of the lattice image 8 is obtained, and orthogonal to the lattice line of the optical system 2 of the magnetic recording track inspection apparatus 20. By obtaining the directional distortion distribution over the entire inspection area 9, the distortion distribution of the portion corresponding to the inspection area 9 of the image obtained in the magnetic recording track inspection apparatus 20 can be calculated.
Further, the CCD camera 3 images the track pattern recorded on the magnetic tape, and calculates the displacement distribution of the track pattern corrected using the distortion distribution, so that the magnetic recording track inspection apparatus 20 can influence the distortion of the image. It is possible to obtain a test result that eliminates the above.
[0075]
In addition, the magnetic recording track inspection apparatus can be calibrated using a predetermined reference gauge.
Further, since the pitch of the lattice pattern on the reference gauge is equal to the pitch of the same azimuth track on the magnetic tape, the luminance distribution of the obtained lattice image becomes close to the luminance distribution of the track pattern image. Therefore, calibration can be executed under conditions close to actual measurement.
In addition, it is possible to perform measurement accuracy confirmation after calibration work with higher accuracy using a predetermined reference gauge for measurement accuracy confirmation.
[0076]
Example 4
Hereinafter, as a fourth embodiment, a case where phase analysis using Fourier transform is used for calibration of an optical system of a magnetic recording track inspection apparatus will be described.
The apparatus used in the fourth embodiment is the same as the magnetic recording track inspection apparatus 20 used in the third embodiment. The procedure of the optical system calibration method of the magnetic recording track inspection apparatus is also the same as the procedure of the optical system calibration method of the magnetic recording track inspection apparatus of the third embodiment. In this embodiment, phase information processing using Fourier transform is used in the calculation of the number of lattices (step 203) in FIG. This will be described.
In the arbitrary region 10 in the inspection region 9 in the lattice image 8 stored in the image memory 4 of FIG. 9, the number of lattices included for each pixel column in the Y direction is calculated by the arithmetic unit 5. The arithmetic device 5 uses a method of phase information processing of the lattice image 8 using Fourier transform described below. FIG. 13 is also an example of a waveform of the luminance distribution of one line having a direction orthogonal to the grid line in the X direction or the Y direction in the inspection region 9 of the grid image 8. If a Fourier transform is applied to this, a frequency spectrum is obtained.
[0077]
FIG. 14 is a schematic diagram of a power spectrum that is a sum of squares of a real part and an imaginary part of a frequency spectrum obtained when the luminance distribution of FIG. 13 is subjected to Fourier transform. When only the primary frequency component (corresponding to the shaded portion in FIG. 14) representing the primary harmonic component of the original waveform is extracted from this frequency spectrum and inverse Fourier transform is performed, the original waveform is smoothed in the real part. The waveform is replaced, and the imaginary part is obtained by shifting the waveform of the real part by half a wavelength. If the arctangent of the imaginary part divided by the real part is taken, the phase value of the first harmonic wave of the original waveform at each pixel is obtained. Of the phase value distribution in the inspection region 9 of the lattice image 8 obtained in this way, the amount of change in the phase value in the arbitrary region 10 is calculated. Since the phase value change amount is 2π and corresponds to one lattice, if the phase change amount is divided by 2π, the number of lattices included in the arbitrary region 10 can be calculated with precision below the decimal point. Therefore, it is possible to obtain a value with higher accuracy than a method using binarization processing.
[0078]
Next, a case where phase information processing using Fourier transform is used in the distortion distribution calculation (step 208) in FIG. 11 will be described. 9, the displacement distribution in the inspection region 9 of the lattice image 8, that is, the optical system 2, is calculated using the actual space length per pixel obtained in the actual space length calculation per pixel (step 206). Calculate the resulting image distortion distribution. If the phase calculation in the inspection area 9 of the lattice image 8 has already been performed at the time of calculating the number of lattices (step 203), the calculation result is used. If the phase calculation is not performed, this is performed in the procedure described in the case of using the phase information processing using Fourier transform in the lattice number calculation (step 203), and the phase value distribution in the inspection region 9 of the lattice image 8 is performed. Is calculated in advance. By taking the difference between the position distribution of each pixel calculated from the actual space length per pixel and the obtained phase value distribution divided by 2π and multiplying by the grid pitch length in each pixel, The displacement distribution in the direction orthogonal to the grid lines is calculated. In the arithmetic processing using the Fourier transform described above, the displacement data is obtained at all the pixel points in the inspection region 9 and no interpolation processing is required. Further, since the position distribution can be calculated with an accuracy of a pixel unit or less, it is possible to accurately calculate the distortion when the distortion is 1 pixel or less. This is effective when the number of lattices is determined as in the case of the magnetic recording track inspection apparatus and the measurement accuracy cannot be improved by increasing the number of lattices.
[0079]
As described above, according to the fourth embodiment, since the number of lattices can be obtained with high accuracy up to a decimal unit by using phase information processing using Fourier transform at the time of calculating the number of lattices, the real space length per pixel can be calculated. Measurement, imaging range calculation, and adjustment can be performed with higher accuracy.
In addition, since the displacement distribution can be calculated in units of decimal pixels by using phase information processing using Fourier transform when calculating the distortion distribution, the distortion distribution is calculated and the displacement distribution is corrected using the distortion distribution with higher accuracy. be able to.
[0080]
【The invention's effect】
The present invention has the following effects.
[0081]
In an arbitrary area within the inspection area of the grid image, the average number of grids for each pixel column of the imaging means is calculated, and the real space length of the arbitrary area in a predetermined direction is calculated by multiplying the grid pitch, By dividing by the number of pixels in the direction, the real space length per pixel of the grid image could be obtained. As a result, calculation of the imaging range, adjustment of the imaging range, calculation of the distortion distribution of the obtained image, correction of the image using the same, and the like can be performed, so the optical system is calibrated to accurately measure the object. Will be able to. Further, these processes can be performed by a series of processes from the capturing of the lattice image.
[0082]
In addition, by using the present invention for a magnetic recording track inspection apparatus, the average real space length and imaging range per pixel in the apparatus, adjustment of the imaging range, calculation of the distortion distribution of the obtained image, The measurement result used can be corrected. Further, these processes can be performed by a series of processes from the capturing of the lattice image.
[0083]
Further, by using a lattice pattern as a reference gauge, Fourier transform processing can be used, and the number of lattices and displacement distribution can be calculated with a precision below the decimal point. Accordingly, the calculation of the real space length and the imaging range per pixel, the adjustment of the imaging range, the calculation of the distortion distribution of the obtained image, and the correction using the same can be performed with higher accuracy.
[0084]
Further, by preparing a reference gauge including a pseudo track pattern, it is possible to check the measurement accuracy with higher accuracy after the optical system is calibrated.
[0085]
In addition, by providing a plurality of types of patterns including pseudo track patterns in one reference gauge, it is possible to calculate the actual space length per pixel, calculate and adjust the imaging range, calculate the distortion distribution, and check the measurement accuracy. This can be done without changing the gauge.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of an optical system calibration system according to an embodiment of the present invention.
FIG. 2 is a diagram showing an example of a lattice pattern in an embodiment of the present invention.
FIG. 3 is a functional block diagram of an optical system calibration method according to an embodiment of the present invention.
FIG. 4 is a schematic diagram of a lattice image in an embodiment of the present invention.
FIG. 5 is an explanatory diagram of thinning processing in the embodiment of the present invention.
FIG. 6 is a diagram showing an example of a lattice pattern in an embodiment of the present invention.
FIG. 7 is a diagram showing an example of a lattice pattern in an embodiment of the present invention.
FIG. 8 is an explanatory diagram of an image correction method according to the embodiment of the present invention.
FIG. 9 is a configuration diagram of a magnetic recording track inspection apparatus according to an embodiment of the present invention.
FIG. 10 is a diagram showing a pattern on a reference gauge in the embodiment of the present invention.
FIG. 11 is a functional block diagram of an optical system calibration method of the magnetic recording track inspection apparatus according to the embodiment of the present invention.
FIG. 12 is a diagram showing an example of a track pattern image of a magnetic tape in an embodiment of the present invention.
FIG. 13 is a diagram showing an example of luminance distribution of one line with a grid image in the embodiment of the present invention.
FIG. 14 is a diagram showing an example of a power spectrum as a result of performing a Fourier transform on a luminance distribution of one line having a lattice image in the embodiment of the present invention.
FIG. 15 is an explanatory diagram of a track pattern displacement distribution correction method according to an embodiment of the present invention.
FIG. 16 is a view showing a reference gauge for checking measurement accuracy in an example of the present invention.
FIG. 17 is a diagram showing an example of a pattern on a reference gauge in the embodiment of the present invention.
FIG. 18 is a configuration diagram of a general image processing system.
[Explanation of symbols]
1 Standard gauge
2 Optical system
3 CCD camera
4 Image memory
5 Arithmetic unit
7 Lattice pattern
7a grid line
8 Grid images
9 Inspection area
10 Arbitrary area
20 Magnetic recording track inspection device
21 Standard gauge
41 line pattern
42 lattice pattern
43 Lattice pattern
44 Reference gauge for measuring accuracy
45 Pseudo track pattern

Claims (25)

An image is formed by imaging an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch, with a direction parallel to and / or perpendicular to the lattice lines as a horizontal scanning direction. Using a first step to obtain a grid image by imaging,
A second step of calculating, for each pixel column, the number of grids corresponding to one pixel column of the imaging unit in a direction orthogonal to the grid line in an arbitrary region in the inspection region of the grid image;
A third step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the second step;
A fourth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice lines by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 5 steps,
A real space length measuring method using an imaging means. The second step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The real space length measuring method by the imaging means of Claim 1 characterized by the above-mentioned. A method for calibrating an optical system in a case where a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via an optical system,
A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
A second step of obtaining a lattice image by imaging the lattice pattern using the imaging means;
A third step of calculating, for each pixel column, the number of lattices corresponding to one pixel column of the imaging means in a direction orthogonal to the lattice line in an arbitrary region in the inspection region in the lattice image;
A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
An optical system calibration method comprising: The third step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The optical system calibration method according to claim 3, further comprising : An image is formed by imaging an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch, with a direction parallel to and / or perpendicular to the lattice lines as a horizontal scanning direction. Using a first step to obtain a grid image by imaging,
A second step of calculating, for each pixel column, the number of grids corresponding to one pixel column of the imaging unit in a direction orthogonal to the grid line in an arbitrary region in the inspection region of the grid image;
A third step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the second step;
A fourth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice lines by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 5 steps,
A sixth step of calculating an imaging range in the inspection region by multiplying the real space length per pixel by the number of pixels in a direction orthogonal to the lattice line of the inspection region;
A real space length measuring method using an imaging means. The second step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The real space length measuring method by the imaging means of Claim 5 characterized by the above-mentioned.   The sixth step calculates the imaging range in the inspection area of the imaging means by multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the lattice line of the inspection area. The method of measuring a real space length by an imaging means according to claim 5, further comprising the step of adjusting the magnification of the optical system so that the imaging range becomes a predetermined value. The second step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The sixth step includes:
By multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the grid line of the inspection area, an imaging range in the inspection area of the imaging unit is calculated. Adjusting the magnification of the optical system to be a value,
The real space length measuring method by the imaging means according to claim 5. A method for calibrating an optical system in a case where a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process is imaged by an imaging means via an optical system,
A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
A second step of obtaining a lattice image by imaging the lattice pattern using the imaging means;
A third step of calculating, for each pixel column, the number of lattices corresponding to one pixel column of the imaging means in a direction orthogonal to the lattice line in an arbitrary region in the inspection region in the lattice image;
A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
A seventh step of calculating an imaging range in the inspection region by multiplying the real space length per pixel by the number of pixels in a direction orthogonal to the grid line of the inspection region;
An optical system calibration method comprising: The third step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The optical system calibration method according to claim 9, further comprising : The seventh step calculates the imaging range in the inspection area of the imaging means by multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the lattice line of the inspection area. 10. The optical system calibration method according to claim 9, further comprising the step of adjusting the magnification of the optical system so that the imaging range becomes a predetermined value. The third step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
And the seventh step comprises:
By multiplying the real space length per pixel by the number of pixels in the direction orthogonal to the grid line of the inspection area, an imaging range in the inspection area of the imaging unit is calculated. Adjusting the magnification of the optical system to be a value,
The optical system calibration method according to claim 9. An image is formed by imaging an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch, with a direction parallel to and / or perpendicular to the lattice lines as a horizontal scanning direction. Using a first step to obtain a grid image by imaging,
A second step of calculating, for each pixel column, the number of grids corresponding to one pixel column of the imaging unit in a direction orthogonal to the grid line in an arbitrary region in the inspection region of the grid image;
A third step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the second step;
A fourth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice lines by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 5 steps,
By using the real space length per pixel to obtain a displacement distribution of the lattice image in a direction perpendicular to the lattice line of the inspection region, the distortion distribution in the direction of the image caused by the optical system is obtained. A sixth step of obtaining the entire inspection area;
An optical system calibration method comprising: The second step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The sixth step includes
Calculating a displacement distribution in a direction perpendicular to the grid line in the inspection region using the real space length per pixel and the phase value distribution which is position information of the grid line in each pixel ; Including,
The optical system calibration method according to claim 13. This is an optical system calibration method in a magnetic recording track inspection apparatus that images and inspects a track pattern of a magnetic recording track on a magnetic tape that has been recorded and visualized by a magnetic recording / reproducing apparatus by an imaging means via an optical system. And
A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
A second step of obtaining a lattice image by imaging the lattice pattern using the imaging means;
In an arbitrary region within the inspection area of the magnetic recording track inspection apparatus in the lattice image, the number of lattices corresponding to one pixel column of the imaging unit in a direction orthogonal to the lattice line is calculated for each pixel column. A third step;
A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
By using the real space length per pixel to obtain a displacement distribution in the direction perpendicular to the grid line in the inspection area of the grid image, the distortion in the direction of the optical system of the magnetic recording track inspection apparatus is obtained. A seventh step of obtaining a distribution over the entire inspection area;
An optical system calibration method for imaging a magnetic recording track. The third step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
And the seventh step comprises:
Using the real space length per pixel and the phase value distribution which is position information of the grid line at each pixel, the displacement distribution of the grid image in the direction perpendicular to the grid line in the inspection region Including the step of calculating
The optical system calibration method according to claim 15. An image is formed by imaging an image obtained by forming an image of a lattice pattern including lattice lines arranged at equal intervals with a known value as a lattice pitch, with a direction parallel to and / or perpendicular to the lattice lines as a horizontal scanning direction. Using a first step to obtain a grid image by imaging,
A second step of calculating, for each pixel column, the number of grids corresponding to one pixel column of the imaging unit in a direction orthogonal to the grid line in an arbitrary region in the inspection region of the grid image;
A third step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the second step;
A fourth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice lines by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 5 steps,
By using the real space length per pixel to obtain a displacement distribution of the lattice image in a direction perpendicular to the lattice line of the inspection region, the distortion distribution in the direction of the optical system is obtained over the entire inspection region. A sixth step obtained in
A seventh step of correcting an image formed by the optical system using the distortion distribution;
An optical system calibration method comprising: The second step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
The sixth step includes:
Calculating a displacement distribution in a direction perpendicular to the grid line in the inspection region using the real space length per pixel and the phase value distribution which is position information of the grid line in each pixel ; Including,
The optical system calibration method according to claim 17. This is an optical system calibration method in a magnetic recording track inspection apparatus that images and inspects a track pattern of a magnetic recording track on a magnetic tape that has been recorded and visualized by a magnetic recording / reproducing apparatus by an imaging means via an optical system. And
A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
A second step of obtaining a lattice image by imaging the lattice pattern using the imaging means;
In an arbitrary region within the inspection area of the magnetic recording track inspection apparatus in the lattice image, the number of lattices corresponding to one pixel column of the imaging unit in a direction orthogonal to the lattice line is calculated for each pixel column. A third step;
A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, a real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
By using the real space length per pixel to obtain a displacement distribution in the direction perpendicular to the grid line in the inspection area of the grid image, the distortion in the direction of the optical system of the magnetic recording track inspection apparatus is obtained. A seventh step of obtaining a distribution over the entire inspection area;
An eighth step of imaging the track pattern recorded on the magnetic tape by the imaging means and calculating a displacement distribution of the track pattern corrected using the distortion distribution;
An optical system calibration method for imaging a magnetic recording track. The third step includes
In the inspection region, performing a Fourier transform in a direction orthogonal to the lattice lines of the lattice pattern, and extracting a primary frequency component from the obtained frequency spectrum;
Performing an inverse Fourier transform on the extracted primary frequency component, and calculating a phase value distribution of the lattice image from a ratio of a real part and an imaginary part of the result;
Calculating the number of grids of each pixel column in a direction orthogonal to the grid lines in an arbitrary area in the inspection area using the phase value distribution;
And the seventh step comprises:
Using the real space length per pixel and the phase value distribution which is position information of the grid line at each pixel, the displacement distribution of the grid image in the direction perpendicular to the grid line in the inspection region Including the step of calculating
The optical system calibration method according to claim 19. The lattice pattern in the first step is a reference gauge for the magnetic recording track inspection device, and the reference gauge is
A line pattern serving as a reference of the coordinate system of the reference gauge on the surface of the reference gauge;
A lattice pattern drawn at an equal pitch so as to have a predetermined angle with respect to a reference direction of the line pattern,
The surface of the line pattern and the lattice pattern when installed in the magnetic recording track inspection apparatus is substantially the same as the height position of the upper surface of the magnetic tape when the magnetic tape is installed in the magnetic recording track inspection apparatus. 20. The method of calibrating an optical system according to claim 3, 9, 15, or 19, wherein the optical systems have equal thicknesses. The optical system calibration method according to claim 21, wherein the pitch of the lattice pattern on the reference gauge is substantially equal to the pitch of the track pattern on the magnetic tape. A method of calibrating a magnetic recording track inspection apparatus for imaging and inspecting a track pattern of a magnetic recording track on a magnetic tape recorded by a magnetic recording / reproducing apparatus and subjected to a visualization process by an imaging means via an optical system,
A grid pattern including grid lines arranged at equal intervals with a known value as a grid pitch and including a grid line parallel to and / or perpendicular to the horizontal scanning direction of the imaging means is installed at a position substantially the same as the imaging position of the magnetic tape. A first step to:
A second scan for obtaining a grid image by imaging the grid pattern using the imaging means. Tep,
In an arbitrary region in the inspection area of the magnetic recording track inspection apparatus in the lattice image, the number of lattices corresponding to one pixel column of the imaging unit in a direction orthogonal to the lattice line is calculated for each pixel column. A third step;
A fourth step of calculating an average number of lattices by averaging the number of lattices for each pixel column obtained in the third step;
A fifth step of calculating a real space length of the arbitrary region in a direction orthogonal to the lattice line by multiplying the average number of lattices by the lattice pitch;
By dividing the real space length by the number of pixels in the direction orthogonal to the grid line in the arbitrary region, the real space length per pixel in the direction orthogonal to the grid line of the grid image is calculated. 6 steps,
By using the real space length per pixel to obtain a displacement distribution in the direction perpendicular to the grid lines in the inspection area of the grid image, the distortion in the direction of the optical system of the magnetic recording track inspection apparatus is obtained. A seventh step of obtaining a distribution over the entire inspection area;
Of the line pattern serving as a reference for the coordinate system and the ideal track pattern recorded on the magnetic tape by two or more heads each having a predetermined azimuth angle in the magnetic recording / reproducing apparatus, one azimuth angle is determined. A track pattern recorded by a head having a light portion and a track pattern recorded by a head having the other azimuth angle is a dark portion, and is substantially equal to the track pattern obtained by at least one displacement. A pseudo track pattern having a known master displacement distribution with respect to the ideal track pattern along the measurement line, and a surface of the pseudo track pattern when the magnetic tape is installed in the magnetic recording track inspection apparatus. An eighth step of installing so as to be substantially equal to the height position;
A ninth step of photographing the pseudo track pattern by the magnetic recording track inspection device and calculating a displacement distribution of the pseudo track pattern corrected using the distortion distribution;
A tenth step of detecting the measurement accuracy of the magnetic recording track inspection apparatus by comparing the displacement distribution and the master displacement distribution;
A method for calibrating a magnetic recording track inspection apparatus, comprising: In calculating the displacement distribution in the ninth step,
In the image of the pseudo track pattern, Fourier transform is performed in the displacement distribution measurement direction, and a primary frequency component is extracted from the obtained frequency spectrum,
Inverse Fourier transform is performed on the extracted primary frequency component, and the phase value distribution of the pseudo track pattern image is calculated from the ratio of the real part and the imaginary part of the result,
Using the phase value distribution, a displacement distribution of the pseudo track pattern image is calculated.
24. The method of calibrating a magnetic recording track inspection apparatus according to claim 23 . The lattice pattern in the first step is a reference gauge for the magnetic recording track inspection device,
The reference gauge is a line pattern serving as a reference of the coordinate system of the reference gauge on the surface of the reference gauge;
A lattice pattern drawn at an equal pitch so as to have a predetermined angle with respect to a reference direction of the line pattern,
In the magnetic recording / reproducing apparatus, among the ideal track patterns recorded on the magnetic tape by two or more heads each having a predetermined azimuth angle, the track pattern recorded by the head having one azimuth angle is clarified. And a pseudo track pattern substantially equal to the track pattern obtained when the track pattern recorded by the head having the other azimuth angle is a dark portion,
Wherein along at least one or more displacement measurement line on the pseudo track pattern, the displacement distribution for the ideal track pattern of the pseudo track pattern is known,
The surface of the line pattern, the lattice pattern, and the pseudo track pattern when the magnetic tape is installed in the magnetic recording track inspection apparatus is the upper surface of the magnetic tape when the magnetic tape is installed in the magnetic recording track inspection apparatus. 20. The method of calibrating an optical system according to claim 3, 9, 15, or 19, wherein a reference gauge having a thickness substantially equal to a height position of the optical system is used.

Images