JP5334265B2 - Chromatic aberration / image distortion correcting apparatus and program therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a magnification chromatic aberration/image distortion correction device and its program which can exactly correct using a simple configuration. <P>SOLUTION: The magnification chromatic aberration/image distortion correction device 1 is provided with: a correction data creation means 2; a correction data memory means 6 which memorizes the correction data; a correction data thinning means 3 which calculates the number of the memory of the correction data based on a memory capacity of the correction data memory means 6 or a predetermined evaluation value, and writes in the correction data memory means 6 after carrying out thinning of input correction data under the number of the memory while a plurality of pieces of the correction data are input; a signal input means 4; a lens parameter input means 5; the correction data memory means 6; a correction data selection means 7; a correction amount calculation means 8; a chromatic difference correction means 9; a signal memory means 10; and a signal output means 11. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a lateral chromatic aberration / image distortion correction apparatus for correcting image color shift (bleeding) caused by lens chromatic aberration (magnification chromatic aberration) and image distortion, and a program thereof.

  In general, the lens of an imaging device (camera) that captures an image has a refractive index that differs depending on the frequency of light, and the position where the image is formed between light having a long wavelength and light having a short wavelength is shifted, so the image is taken through the lens. Blurred image is produced. In an image composed of the three primary colors of red (R), green (G), and blue (B), red having a longer wavelength and blue having a shorter wavelength than green are shifted in different directions. Red or blue blurring occurs at the border with white.

  Therefore, the amount of deviation of the imaging position of red and blue light relative to green due to this chromatic aberration of magnification is obtained in advance for each pixel, and by correcting the deviation for the red and blue primary color signals of the captured image, A technique for correcting image blur due to lateral chromatic aberration is disclosed (see Patent Document 1). Also disclosed is a technique for converting luminance signals and color difference signals into primary color signals of red, green, and blue, and correcting distortion due to lens distortion and blur due to lateral chromatic aberration (see Patent Document 2).

  The inventions described in Patent Document 1 and Patent Document 2 can be realized when lens parameters such as zoom, focus, and iris are constant. On the other hand, when shooting a video (moving image) such as a TV program with a camera, generally, the lens parameter is often changed in various ways. In this case, in the inventions described in Patent Document 1 and Patent Document 2, if data indicating the amount of deviation of the imaging position corresponding to the change of the lens parameter is prepared in advance, it is difficult to acquire this data. In addition, a huge amount of data needs to be stored, which is not realistic.

  In order to solve this problem, for example, the invention described in Patent Document 3 has been proposed. The invention described in Patent Document 3 eliminates blurring of an image due to chromatic aberration of magnification based on the lens parameter even when the lens parameter is changed to various values.

JP-A-6-292207 (paragraph numbers 0012 to 0030, FIGS. 1 to 5) JP 2000-3437 (paragraph numbers 0033 to 0070, FIGS. 3 to 7) JP 2006-135805 A (paragraph numbers 0022 to 0052, FIGS. 1 to 4)

  However, the invention described in Patent Document 3 does not indicate an appropriate standard for the number of correction data stored in the storage unit. For this reason, the invention described in Patent Document 3 requires a large-capacity storage unit when the number of correction data is too large. On the other hand, the invention described in Patent Document 3 makes it difficult to select appropriate correction data if the number of correction data is too small, and the correction amount calculated using the correction data is significantly different from the actual shift amount of the video. Therefore, the correction cannot be performed accurately.

  Accordingly, an object of the present invention is to solve the above-described problems and to provide a magnification chromatic aberration / image distortion correction apparatus and a program thereof capable of performing accurate correction with a simple configuration.

  In order to solve the above-described problem, the lateral chromatic aberration / image distortion correction apparatus according to the first invention of the present application is based on the value of a lens parameter of a camera that captures an image, and an image formed by the lateral chromatic aberration of each pixel of the image in the lens parameter. A chromatic aberration of magnification that corrects at least one of the color shift caused by the chromatic aberration of magnification and the distortion of the image, using correction data that associates the amount of positional deviation and the aberration amount indicating at least one of the amount of distortion of the image. An image distortion correction apparatus, wherein correction data storage means, correction data thinning means, signal input means, lens parameter input means, correction data selection means, correction amount calculation means, correction means, and signal output means It is characterized by providing.

According to this configuration, the lateral chromatic aberration / image distortion correction device stores the correction data of the first primary color, the second primary color, and the third primary color, which are the three primary colors, by the correction data storage unit. Further, the magnification chromatic aberration / image distortion correction device receives a plurality of the correction data by the correction data thinning-out means, and determines the storage number of the correction data based on the storage capacity of the correction data storage means or a predetermined evaluation value. The calculated correction data is thinned below the stored number, and then written into the correction data storage means.
Thus, the magnification chromatic aberration / image distortion correction apparatus can store an appropriate number of correction data in the correction data storage means. Note that it is preferable that the magnification chromatic aberration / image distortion correction apparatus stores correction data for different lens parameter values in the correction data storage means before correcting the image.

  Further, the magnification chromatic aberration / image distortion correction device uses a signal input means to provide a first primary color signal, a second primary color signal, and a third primary color signal indicating the brightness of the first primary color, the second primary color, and the third primary color for each pixel. Input the primary color signal. Further, the magnification chromatic aberration / image distortion correcting device has the image corresponding to each pixel of the first primary color signal, the second primary color signal, and the third primary color signal input from the signal input unit by the lens parameter input unit. A shooting lens parameter that is a lens parameter when shooting is input. The magnification chromatic aberration / image distortion correction device uses the correction data selection unit to calculate the value of the photographic lens parameter stored in the correction data storage unit based on the photographic lens parameter input from the lens parameter input unit. Select correction data for the two lens parameters to be sandwiched. Here, the magnification chromatic aberration / image distortion correction apparatus corrects lens parameter correction data having a value larger than the photographing lens parameter value and lens parameter having a smaller value among the correction data stored in the correction data storage means. Select data.

  Further, the magnification chromatic aberration / image distortion correction device is configured to correct at least the color shift due to the magnification chromatic aberration and the distortion of the image in the photographing lens parameter based on the correction data selected by the correction data selection unit by the correction amount calculation unit. A correction amount for correcting one is calculated by interpolation. Then, the magnification chromatic aberration / image distortion correction device uses the correction unit to determine the brightness of each pixel of the first primary color signal, the second primary color signal, and the third primary color signal based on the correction amount calculated by the correction amount calculation unit. The pixel value indicating the height is corrected. Further, the magnification chromatic aberration / image distortion correction apparatus outputs the first primary color signal, the second primary color signal, and the third primary color signal, in which the pixel value is corrected by the correction unit, by the signal output unit.

Further, in the magnification chromatic aberration / image distortion correction device according to the second invention of the present application, the correction data thinning-out means is based on a value obtained by dividing the storage capacity of the correction data storage means by the capacity per correction data. The number of memories is calculated.
According to such a configuration, the magnification chromatic aberration / image distortion correction apparatus can thin out the correction data so as not to exceed the storage capacity of the correction data storage unit even when a large amount of correction data is input.

  Further, in the magnification chromatic aberration / image distortion correction device according to the third invention of the present application, the correction data thinning means interpolates the correction data of the lens parameter value and the thinning candidate correction data for each pixel and the correction data. The difference between the aberration amounts is calculated as the evaluation value, and the maximum value or the average value of the differences is calculated as the evaluation value for each value of the lens parameter. The maximum number of candidate correction data is calculated as the stored number.

  Here, as the number of thinning out correction data is increased, the storage capacity of the correction data storage means can be saved. On the other hand, when the thinned correction data is interpolated, it is considered that an error with respect to the correction data before thinning (that is, the generated correction data) increases. For this reason, the lateral chromatic aberration / image distortion correction device interpolates the post-interpolation correction data from the correction data of the lens parameter that is a thinning candidate by the same method as the correction amount calculation means. Then, the magnification chromatic aberration / image distortion correction apparatus calculates an evaluation value indicating an error of the post-interpolation correction data based on a difference in aberration amount between the correction data and the post-interpolation correction data. Further, the magnification chromatic aberration / image distortion correction apparatus calculates the number of memories so that the error (evaluation value) of the correction data after interpolation is within an allowable range and the storage capacity of the correction data storage means can be saved most.

  In order to solve the above-described problem, the lateral chromatic aberration / image distortion correction program according to the fourth invention of the present application is based on the value of a lens parameter of a camera that captures an image and the lateral chromatic aberration of each pixel of the image in the lens parameter. In order to correct at least one of the color shift due to the chromatic aberration of magnification and the distortion of the image by using correction data in which an amount of deviation of the imaging position and an aberration amount indicating at least one of the distortion amount of the image are associated with each other A correction data storage means for storing the correction data of the first primary color, the second primary color and the third primary color, which are the three primary colors, a correction data thinning means, a signal input means, a lens parameter input means, and correction data; It functions as selection means, correction amount calculation means, correction means, and signal output means.

According to such a configuration, the magnification chromatic aberration / image distortion correction program receives a plurality of the correction data by the correction data thinning-out means, and the correction data based on the storage capacity of the correction data storage means or a predetermined evaluation value. Is stored in the correction data storage means after the input correction data is thinned below the storage number.
Thus, the magnification chromatic aberration / image distortion correction program can store an appropriate number of correction data in the correction data storage means. Note that the computer executing the magnification chromatic aberration / image distortion correction program preferably stores correction data for different lens parameter values in the correction data storage means before correcting the image.

  In addition, the lateral chromatic aberration / image distortion correction program uses a signal input unit to output a first primary color signal, a second primary color signal, and a third primary color signal indicating the brightness of the first primary color, the second primary color, and the third primary color for each pixel. Input the primary color signal. In addition, the magnification chromatic aberration / image distortion correction program stores the image corresponding to each pixel of the first primary color signal, the second primary color signal, and the third primary color signal input from the signal input unit by the lens parameter input unit. A shooting lens parameter that is a lens parameter when shooting is input. The magnification chromatic aberration / image distortion correction program calculates the value of the photographic lens parameter stored in the correction data storage unit based on the photographic lens parameter input from the lens parameter input unit by the correction data selection unit. Select correction data for the two lens parameters to be sandwiched. Here, the lateral chromatic aberration / image distortion correction program is a correction data stored in the correction data storage means for correcting the lens parameter having a value larger than the value of the photographing lens parameter and correcting the lens parameter having a smaller value. Select data.

  Further, the magnification chromatic aberration / image distortion correction program, based on the correction data selected by the correction data selection means by the correction amount calculation means, at least color shift due to the magnification chromatic aberration and distortion of the image in the photographing lens parameter. A correction amount for correcting one is calculated by interpolation. Then, the magnification chromatic aberration / image distortion correction program causes the brightness of each pixel of the first primary color signal, the second primary color signal, and the third primary color signal to be corrected by the correction unit based on the correction amount calculated by the correction amount calculation unit. The pixel value indicating the height is corrected. Further, the magnification chromatic aberration / image distortion correction program outputs the first primary color signal, the second primary color signal, and the third primary color signal whose pixel values have been corrected by the correction unit by the signal output unit.

The present invention has the following excellent effects.
In the first and fourth inventions of the present application, since an appropriate number of correction data is stored in the correction data storage means, the number of correction data stored in the correction data storage means is neither too much nor too little. Thus, the first and fourth inventions of the present application enable accurate correction, save the storage capacity of the correction data storage means, and simplify the configuration of the magnification chromatic aberration / image distortion correction apparatus.

  In the second invention of the present application, even when a large amount of correction data is input, the correction data can be thinned out so as not to exceed the storage capacity of the correction data storage means, and a large-scale correction data storage means is not required. Therefore, the configuration of the magnification chromatic aberration / image distortion correction apparatus can be simplified.

  Since the third invention of the present application calculates the number of memories in which the error of the correction data after interpolation is within an allowable range and the storage capacity of the correction data storage means can be saved most, the configuration of the magnification chromatic aberration / image distortion correction apparatus can be simplified. In addition, the video can be corrected accurately.

1 is a block diagram showing a configuration of a magnification chromatic aberration / image distortion correction apparatus according to an embodiment of the present invention. It is explanatory drawing for demonstrating the example of the correction data which the correction data production | generation means of FIG. 1 produced | generated. It is explanatory drawing for demonstrating the example of the correction data thinned out by the correction data thinning means of FIG. It is explanatory drawing for demonstrating the example which calculates the difference of the amount of aberrations by the correction data thinning-out means of FIG. FIG. 2 is an explanatory diagram for explaining a method in which the aberration correction unit in FIG. 1 calculates a pixel value by shifting a certain pixel based on a correction amount, and (a) is a diagram for explaining a process of a shift processing unit. (B) is a figure explaining the process of a two-dimensional filter part. 2 is a flowchart showing an operation of generating correction data by the correction data generating means of FIG. 2 is a flowchart showing an operation in which correction data thinning means in FIG. 1 thins correction data. In embodiment of this invention, it is explanatory drawing for demonstrating the example which uses zoom, a focus, and an iris as a lens parameter. 2 is a flowchart illustrating an operation of correcting an image by the magnification chromatic aberration correction apparatus of FIG. 1.

(Outline of the Invention)
The outline of the present invention will be briefly described below with reference to FIG.
The lateral chromatic aberration / image distortion correcting device 1 corrects color shift due to lateral chromatic aberration in an image and distortion of the image such as a barrel shape or an inverted barrel shape using correction data described later. Here, the lateral chromatic aberration / image distortion correction apparatus 1 stores correction data for different lens parameter values in the correction data storage means 6 before each image is corrected by each means within the broken line in FIG. There is a need. Therefore, the magnification chromatic aberration / image distortion correction apparatus 1 shoots a predetermined pattern while changing lens parameters of a camera lens (not shown), measures color shift and image distortion due to magnification chromatic aberration, and corrects the correction data. Is generated. Then, the magnification chromatic aberration / image distortion correction apparatus 1 thins out the generated correction data, and then stores it in the correction data storage means 6. Thereafter, the magnification chromatic aberration / image distortion correction apparatus 1 actually corrects the input image using the correction data stored in the correction data storage means 6.

  Although the type of the camera is not particularly limited, it will be described here as a three-plate broadcast program shooting camera including a spectral means such as a prism and shooting elements for each of the three primary colors. That is, the camera outputs the video of the broadcast program to the magnification chromatic aberration / image distortion correction device 1 (signal input means 4).

(Embodiment)
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate. In each embodiment, means having the same function are denoted by the same reference numerals and description thereof is omitted.
As shown in FIG. 1, the magnification chromatic aberration / image distortion correction apparatus 1 includes a correction data generation unit 2, a correction data thinning unit 3, a signal input unit 4, a lens parameter input unit 5, and a correction data storage unit 6. Correction data selection means 7, correction amount calculation means 8, aberration correction means (correction means) 9, signal storage means 10, and signal output means 11.

  The correction data generation means 2 (2a, 2b, 2c) receives, for example, a reference signal (G signal, B signal or R signal), a comparison signal (undistorted ideal signal or processed G signal), and a lens parameter. Is done. Here, the correction data generating means 2a receives the G signal as the reference signal and the undistorted ideal signal as the comparison signal. Further, the correction data generation means 2b receives the B signal as a reference signal and the processed G signal (undistorted G signal) as a comparison signal. Further, the correction data generating means 2c receives the R signal as a reference signal and the processed G signal (undistorted G signal) as a comparison signal.

  Then, the correction data generation means 2 (2a, 2b, 2c) generates correction data using the inputted reference signal, comparison signal, and lens parameter. Further, the correction data generating means 2 (2a, 2b, 2c) outputs the generated correction data to the correction data thinning means 3 (3a, 3b, 3c).

  Here, the lens parameters are zoom, focus, iris and the like of the camera lens. Here, the correction data generation means 2 (2a, 2b, 2c) may input only one lens parameter of zoom, focus, and iris, for example, zoom, or may select a plurality of types of lens parameters. It is good also as inputting. Here, the correction data generation means 2 (2a, 2b, 2c) inputs only zoom as a lens parameter.

Hereinafter, an example of correction data generated by the correction data generation means 2 (2a, 2b, 2c) will be described with reference to FIG. 2 (see FIG. 1 as appropriate).
In FIG. 2, the X axis is the zoom (lens parameter), the Y axis is the pixel aberration amount (the amount of image position shift and the amount of distortion), and the Z axis is the distance from the image center (number of pixels). . That is, the correction data is obtained by associating the value of the lens parameter with the aberration amount for each pixel of the G signal, B signal, or R signal.

  Here, although the correction data generating means 2 (2a, 2b, 2c) continuously changes the zoom, the time required for generating the correction data and the storage capacity of the correction data storage means 6 (6a, 6b, 6c) Considering the problem, the correction data is generated at a predetermined sample interval.

In the following description, the number of correction data generated at a predetermined sample interval (measurement points) to F _1. That is, the value of the measurement points _{F 1} is two or more.
In the sampled correction data, the m-th zoom counted from the minimum zoom value is z (m).
Further, correction data corresponding to the zoom z (m) is a correction data plane ab (m).
This correction data plane is one correction data corresponding to a certain zoom (lens parameter) value. In the example of FIG. 2, the correction data with a zoom value of 12 millimeters is one correction data plane, and the correction data with a zoom value of 14 millimeters is also one correction data plane.

Hereinafter, returning to FIG. 1, the description of the configuration of the magnification chromatic aberration / image distortion correction apparatus 1 will be continued.
The correction data thinning means 3 (3a, 3b, 3c) receives a plurality of correction data from the correction data generation means 2 (2a, 2b, 2c) and stores the correction data storage means 6 (6a, 6b, 6c). The number of stored correction data is calculated based on the capacity or an evaluation value described later. Then, the correction data thinning means 3 (3a, 3b, 3c) writes the input correction data to the correction data storage means 6 (6a, 6b, 6c) after thinning it below the stored number.

  Hereinafter, with reference to FIG. 3 and FIG. 4, two specific examples of the thinning method based on the storage capacity of the correction data storage means 6 (6a, 6b, 6c) and the storage number calculation method based on the evaluation value will be described in detail. (Refer to FIG. 1 as appropriate).

<First Example: Thinning Method Based on Storage Capacity of Correction Data Storage Unit>
In this first example, the correction data thinning-out means 3 (3a, 3b, 3c) is based on a value obtained by dividing the storage capacity of the correction data storage means 6 (6a, 6b, 6c) by the capacity per correction data plane. Thus, the storage number (number of correction data planes) that can be stored in the correction data storage means 6 (6a, 6b, 6c) is calculated. At this time, the correction data thinning-out means 3 (3a, 3b, 3c) may round off or round off when the divided value has a decimal point.

  Here, for example, as the storage capacity of the correction data storage means 6 (6a, 6b, 6c), the maximum storage capacity of the correction data storage means 6 (6a, 6b, 6c) is set in advance. Further, for example, the capacity per correction data plane may be set in advance, and may be the average data capacity or the maximum data capacity of the correction data planes input to the correction data thinning means 3 (3a, 3b, 3c). Good.

Then, the correction data thinning-out means 3 (3a, 3b, 3c) may thin out the correction data so as to be within the range of the calculated storage number (number of correction data planes). Here, the correction data thinning-out means 3 (3a, 3b, 3c) skips the correction data plane when the zoom is at the minimum value (m = 1) and when the zoom is at the maximum value (m = F ₁ ). Do not draw. Further, the correction data thinning means 3 (3a, 3b, 3c) is not particularly limited as to which thinning out correction data planes other than the minimum value and the maximum value, but it is preferable to thin out the correction data planes at equal intervals.

  In the example of FIG. 3, the correction data plane thinned out by the correction data thinning means 3 (3a, 3b, 3c) is indicated by a broken line. That is, in the example of FIG. 3, the correction data thinning means 3 (3a, 3b, 3c) thins out correction data planes with zooms of 14 millimeters and 16 millimeters.

  According to the first example, the lateral chromatic aberration / image distortion correction device 1 corrects so as not to exceed the storage capacity of the correction data storage means 6 (6a, 6b, 6c) even when a large amount of correction data is input. Data can be thinned out. Therefore, the magnification chromatic aberration / image distortion correction apparatus 1 does not require the large-scale correction data storage means 6 (6a, 6b, 6c), and can simplify the configuration.

<Second example: Thinning method based on evaluation value>
In the second example, as in the first example, the correction data thinning-out means 3 (3a, 3b, 3c) has a minimum zoom value (m = 1) and a maximum zoom value (m = F _1). ) Correction data plane is not thinned out. Therefore, consider a case where one correction data plane is thinned out from F ₁ for correction data planes other than these. At this time, the number of correction data planes that are candidates for thinning is F ₁ -2.

  In this case, the correction data thinning-out means 3 (3a, 3b, 3c) uses the same interpolation method as the correction amount calculation means 8 (8a, 8b, 8c) described later, such as linear interpolation and spline interpolation, and the correction data of the thinning candidates. The interpolation value β is calculated. For example, when thinning out the correction data plane ab (2), the correction data thinning-out means 3 (3a, 3b, 3c) determines a thinning candidate from the aberration amounts of the correction data plane ab (1) and the correction data plane ab (3). Is calculated by interpolating the aberration amount of the correction data plane ab (2) ′.

  For example, in FIG. 3, a correction data plane ab (2) with a zoom of 14 millimeters and a correction data plane ab (3) with a zoom of 16 millimeters are thinning candidates. In this case, the correction data thinning-out means 3 (3a, 3b, 3c), as shown in FIG. 4, measures the aberration amount measurement value β2 of the correction data plane ab (2) and the thinning candidate correction data plane ab (2). The difference between the aberration amount of ′ and the interpolation value β′2 is calculated. Similarly, the correction data thinning-out means 3 (3a, 3b, 3c) determines the aberration amount measurement value β3 of the correction data plane ab (3) and the aberration amount of the thinning candidate correction data plane ab (3) ′. The difference from the interpolation value β′3 is calculated.

  4 is an enlarged view of the correction data plane from 12 millimeters to 18 millimeters when the Z-axis value (distance from the image center) = α in FIG. 3 and viewed from the Z-axis direction. It is an enlarged view. The measured value is the aberration amount of the correction data plane generated by the correction data generation unit 2 (2a, 2b, 2c).

At this time, the correction data thinning-out means 3 (3a, 3b, 3c) calculates the aberration amount difference between the correction data plane and the post-decimation candidate post-correction data plane for all pixels including the Z-axis value ≠ α. . Then, the correction data thinning-out means 3 (3a, 3b, 3c) sets, as an evaluation value candidate, a pixel having the maximum difference in aberration amount among all the pixels. Further, the correction data thinning-out means 3 (3a, 3b, 3c) may use an average value of the aberration amount differences as evaluation value candidates for all the pixels. Here, representing the evaluation value candidates in the case of thinning out the correction data plane number _F 1 -1 in the second correction data plane _{(m = 2) Ev_temp (F} 1 -1,2).

Similarly, the correction data thinning-out means 3 (3a, 3b, 3c) evaluates the evaluation value candidate Ev_temp (F) when the third correction data plane (m = 3) is thinned out by the number of correction data planes F ₁ −1. ₁ 1,3) is calculated. Then, correction data thinning means 3 (3a, 3b, 3c) are sequentially evaluated when the correction data plane number _F 1 -1 thinned out _F 1 -1-th correction data plane _{(m =} F 1 -1) Up to value candidates Ev_temp (F ₁ −1, F ₁ −1) are calculated. Furthermore, correction data thinning means 3 (3a, 3b, 3c), as shown in the following equation (1), the evaluation value candidate _{Ev_temp (F 1 -1,2) ~Ev_temp (} F 1 -1, F 1 - The smallest of 1) is calculated as the evaluation value EV (F ₁ −1) of the number of correction data planes F ₁ −1. Furthermore, the correction data thinning-out means 3 (3a, 3b, 3c) calculates the correction data plane number at this time as a reduced plane number DP (F ₁ -1).

EV (F ₁ −1) = min (Ev_temp (F ₁ −1, t)) Equation (1)
In the formula (1), 2 <t <F ₁ −1. In the formula (1), min is a function for obtaining a minimum value.

Furthermore, the correction data thinning-out means 3 (3a, 3b, 3c) sequentially evaluates the evaluation value EV until two correction data planes are thinned out until F _1-2 correction data planes are thinned out. (F ₁ -2) to EV (2) and reduced plane numbers DP (F ₁ -2) to DP (2) are calculated.

Further, the correction data thinning-out means 3 (3a, 3b, 3c) has the evaluation values EV (F ₁ -2) to EV (2) equal to or less than the maximum allowable color misregistration amount and the number of correction data planes to be thinned out. The number of memories (the number of correction data planes) is calculated so that becomes the maximum. The allowable value of the maximum color misregistration amount is a preset threshold value. For example, the document “M. Kanazawa, et al.,“ Color error from an RGB-stripe pixel structure, ”Journal of the Society for Information Display Tenth Color Imaging Conference: Color Science and Engineering Systems, Technologies, Applications, vol. 11, no. 2, pp. 387-393, 2003 ”.

  Thereafter, the correction data thinning-out means 3 (3a, 3b, 3c) may thin out the correction data plane within the range of the calculated storage number. The correction data thinning means 3 (3a, 3b, 3c) is not particularly limited as to which correction data plane is thinned out as in the first example, but it is preferable to thin out the correction data planes at equal intervals.

  According to the second example, the magnification chromatic aberration / image distortion correction apparatus 1 has an error (evaluation value) of the correction data after interpolation within an allowable range, and is stored in the correction data storage means 6 (6a, 6b, 6c). Calculate the number of memories that can save the most capacity. Therefore, the lateral chromatic aberration / image distortion correction apparatus 1 can accurately correct the image shift while suppressing an increase in the storage capacity of the correction data storage means 6 (6a, 6b, 6c).

  Here, the magnification chromatic aberration / image distortion correction apparatus 1 may set in advance which of the first example and the second example is used. For example, when the correction data storage means 6 (6a, 6b, 6c) cannot be added, or when it is desired to reduce the cost of the magnification chromatic aberration / image distortion correction apparatus 1, the method of the first example is preferable. On the other hand, when importance is attached to the accuracy of video correction, that is, video quality (image quality), the method of the second example is preferable.

Hereinafter, returning to FIG. 1, the description of the configuration of the magnification chromatic aberration / image distortion correction apparatus 1 will be continued.
The signal input means 4 (4a, 4b, 4c) inputs three primary color signals (G signal, B signal and R signal) from the outside. Here, the signal input means 4a inputs a G signal indicating the brightness of the green pixel (G pixel) and outputs it to the aberration correction means 9a. The signal input unit 4b receives a B signal indicating the brightness of the blue pixel (B pixel) and outputs the B signal to the aberration correction unit 9b. Further, the signal input means 4c inputs an R signal indicating the brightness of the red pixel (R pixel) and outputs it to the aberration correction means 9c.

  The lens parameter input means 5 inputs lens parameters of the camera that has captured the image from the outside. Here, the lens parameter input means 5 receives lens parameters (photographing lens parameters) corresponding to the pixels of the video signal input to the signal input means 4 (4a, 4b, 4c). The lens parameter input means 5 outputs the input photographic lens parameters to the correction data selection means 7 (7a, 7b, 7c) and the correction amount calculation means 8 (8a, 8b, 8c). The lens parameter input means 5 uses the same type of lens parameters as the correction data generation means 2 (2a, 2b, 2c).

  In the correction data storage means 6 (6a, 6b, 6c), correction data is written from the correction data thinning means 3 (3a, 3b, 3c). For example, the correction data storage means 6 (6a, 6b, 6c) is a general storage means such as a semiconductor memory or a hard disk.

  Based on the lens parameters input from the lens parameter input unit 5, the correction data selection unit 7 (7a, 7b, 7c) stores the lens parameters stored in the correction data storage unit 6 (6a, 6b, 6c). The lens parameter correction data close to the value is selected. Specifically, the correction data selection means 7 (7a, 7b, 7c) has the closest correction data larger than the zoom value input from the lens parameter input means 5 and the input zoom value. Two correction data of smaller and closest correction data are selected. Then, the correction data selection means 7 (7a, 7b, 7c) outputs the selected correction data to the correction amount calculation means 8 (8a, 8b, 8c). When there are two or more types of lens parameters, the correction amount can be calculated by interpolation in the correction amount calculation means 8 to be described later by selecting correction data that sandwiches the input lens parameter values. it can.

  The correction amount calculation means 8 (8a, 8b, 8c) is based on the lens parameters input from the lens parameter input means 5, and each pixel of the primary color signal input from the signal input means 4 (4a, 4b, 4c). The correction amount indicating the amount of color misregistration due to lateral chromatic aberration and the amount of image distortion is calculated. Then, the correction amount calculation unit 8 (8a, 8b, 8c) outputs the calculated correction amount to the aberration correction unit 9 (9a, 9b, 9c).

Specifically, the correction amount calculation means 8 (8a, 8b, 8c) receives one or more correction data from the correction data selection means 7 (7a, 7b, 7c), and based on this correction data, the lens A correction amount for the photographic lens parameter input from the parameter input means 5 is calculated by interpolation. Here, the correction amount calculation means 8 receives two correction data from the correction data selection means 7 (7a, 7b, 7c) and calculates the correction amount by linear interpolation. In this case, the correction amount Y ₀ of the horizontal direction of the correction amount X ₀ vertical direction is determined by the following formula (2).

_{X 0 = X A + (X} B -X A) (Z 0 -Z A) / (Z B -Z A)
_{Y 0 = Y A + (Y} B -Y A) (Z 0 -Z A) / (Z B -Z A) ··· Equation (2)
In Expression (2), the zoom (photographing lens parameter) value input from the lens parameter input unit 5 is Z ₀ , and the zoom, horizontal movement amount, and vertical movement amount of the two correction data input from the correction data selection unit 7. Are Z _A , Z _B , X _A , X _B , Y _A and Y _B , respectively.

  Here, the case where only the zoom is input as the photographing lens parameter has been described, but the present invention is not limited to this. That is, the correction amount calculation means 8 (8a, 8b, 8c) similarly applies the lens input from the correction data selection means 7 (7a, 7b, 7c) when two or more types of photographing lens parameters are input. Based on the parameter correction data, the correction amount can be calculated by interpolation processing such as linear interpolation and spline interpolation.

  The aberration correction unit 9 (9a, 9b, 9c) is input from the signal input unit 4 (4a, 4b, 4c) based on the correction amount input from the correction amount calculation unit 8 (8a, 8b, 8c). It corrects color shift and image distortion due to lateral chromatic aberration in the G signal, B signal, and R signal. Here, the aberration correction unit 9a corrects the G signal input from the signal input unit 4a, and writes the corrected pixel value of the G signal to the address indicating the corrected pixel position of the signal storage unit 10a. The aberration correction unit 9b corrects the B signal input from the signal input unit 4b, and writes the corrected pixel value of the B signal at an address indicating the corrected pixel position of the signal storage unit 10b. Further, the aberration correction unit 9c corrects the R signal input from the signal input unit 4c, and writes the corrected pixel value of the R signal to the address indicating the corrected pixel position of the signal storage unit 10c. Therefore, the aberration correction unit 9 (9a, 9b, 9c) includes a shift processing unit 91 (91a, 91b, 91c) and a two-dimensional filter unit 92 (92a, 92b, 92c).

Here, with reference to FIG. 5, a method of calculating the pixel value of each pixel in which the color misregistration and the image distortion are corrected by the aberration correcting means 9 (9a, 9b, 9c) will be described.
The shift processing unit 91 (91a, 91b, 91c) extracts an integer component of the correction amount input from the correction amount calculation unit 8 (8a, 8b, 8c), and the signal input unit 4 based on the value of the integer component. The G signal, B signal and R signal input from (4a, 4b, 4c) are shifted. Specifically, the shift processing unit 91 (91a, 91b, 91c), for each pixel, the horizontal correction amount and the vertical correction amount input from the correction amount calculation means 8 (8a, 8b, 8c). Extract integer components from. Then, as shown in FIG. 5A, the shift processing unit 91 (91a, 91b, 91c) specifies a pixel obtained by moving the target pixel by the amount of the integer component. Here, if the integer component of the horizontal correction amount of the pixel P _x , _{y at} the y line position in the horizontal direction is ΔX and the integer component of the vertical correction amount is ΔY, this integer The positions of the pixels P _x ′, _y ′ indicated by the components ΔX, ΔY are (x ′, y ′) = (x + ΔX, y + ΔY).

The two-dimensional filter unit 92 (92a, 92b, 92c) extracts the decimal component of the correction amount input from the correction amount calculation means 8 (8a, 8b, 8c), and shift processing unit based on the value of the decimal component The pixel value of the pixel shifted by 91 (91a, 91b, 91c) is calculated. Specifically, the two-dimensional filter unit 92 (92a, 92b, 92c) has the pixel values of the pixels P _x ′, _y ′ shifted by the amount of the integer component by the shift processing unit 91 (91a, 91b, 91c). Is calculated. That is, the two-dimensional filter unit 92 (92a, 92b, 92c) extracts the decimal component of the correction amount, and further shifts the pixel shifted by the amount of the integer component of the correction amount by the amount of the decimal component. At this time, the pixel value of the pixel is calculated. Here, as shown in FIG. 5B, the fractional components of the horizontal and vertical correction amounts extracted by the two-dimensional filter unit 92 (92a, 92b, 92c) are respectively represented by f1, f2 (horizontal of one pixel). And the vertical size is 1), and the pixel values of the pixel and the pixels adjacent to the upper, left, and upper left of the pixel are Ga, Gb, Gc, and Gd, respectively. In this case, two-dimensional filter portion 92 (92a, 92b, 92c), the pixel _{P x ', y'} pixel values _{G x} of _{', y'} and is calculated by the following equation (3).

_Gx ', _y ' = a * Ga + b * Gb + c * Gc + d * Gd ... Formula (3)
a = (1-f1). (1-f2)
b = (1-f1) · f2
c = f1 · (1-f2)
d = f1 · f2

Then, the two-dimensional filter unit 92 (92a, 92b, 92c) converts the pixel values G _x ′, _y ′ into the shift processing units 91 (91a, 91b, 91c) in the signal storage means 10 (10a, 10b, 10c). ) Is written in the address indicating the position of the pixel subjected to the shift processing.

Returning to FIG. 1, the description will be continued.
The signal storage means 10 (10a, 10b, 10c) is a general storage means such as a semiconductor memory for storing pixel values of a video signal in which color shift due to lateral chromatic aberration and image distortion are corrected. Here, the signal storage means 10a stores the pixel value of the G signal corrected by the aberration correction means 9a. The signal storage unit 10b stores the pixel value of the B signal corrected by the aberration correction unit 9b. Further, the signal storage unit 10c stores the pixel value of the R signal corrected by the aberration correction unit 9c.

  The signal output means 11 (11a, 11b, 11c) reads out the pixel value stored in the signal storage means 10 (10a, 10b, 10c) and outputs it as a video signal with corrected magnification chromatic aberration. Here, the signal output means 11a reads the pixel value of the G signal stored in the signal storage means 10a and outputs it as a corrected G signal. The signal output unit 11b reads out the pixel value of the B signal stored in the signal storage unit 10b and outputs it as a corrected B signal. Further, the signal output means 11c reads out the pixel value of the R signal stored in the signal storage means 10c and outputs it as a corrected R signal.

[Operation of magnification chromatic aberration correction device]
With reference to FIGS. 6 to 9, the operation of the correction chromatic aberration / image distortion correction device 1 for generating correction data, the operation for thinning out correction data, and the operation for correcting video will be described in order (see FIG. 1 as appropriate).

<Correction data generation operation>
As shown in FIG. 6, the lateral chromatic aberration / image distortion correction device 1 extracts feature points from the video indicated by the input reference signal (G signal, B signal, or R signal) by the correction data generation means 2 ( Step S1). Then, the magnification chromatic aberration / image distortion correction apparatus 1 extracts feature points from the video indicated by the input comparison signal (undistorted ideal signal or processed G signal) by the correction data generation means 2 (step S2).

  Further, the magnification chromatic aberration / image distortion correction device 1 searches the corresponding point between the feature point of the reference signal and the feature point of the comparison signal by the correction data generation means 2 (step S3). Then, the magnification chromatic aberration / image distortion correction apparatus 1 calculates an aberration amount based on the corresponding points by the correction data generation unit 2 (step S4).

  Further, the magnification chromatic aberration / image distortion correction device 1 determines whether or not the correction data generation means 2 has calculated this aberration amount for the entire screen (step S5). Here, when the aberration amount is calculated for the entire screen (Yes in step S5), the lateral chromatic aberration / image distortion correction apparatus 1 proceeds to the process of step S6. On the other hand, if the aberration amount has not been calculated for the entire screen (No in step S5), the lateral chromatic aberration / image distortion correction apparatus 1 returns to the process of step S3.

Further, the magnification chromatic aberration / image distortion correction apparatus 1 generates correction data by associating an aberration amount with the input lens parameter by the correction data generation means 2 and uses the correction data as the correction data thinning means 3 (3a, 3b). 3c) (step S6).
In FIG. 6, steps S1 and S2 are parallel processing. However, serial processing may be performed in the order of steps S1 and S2 or in the reverse order.

<Correction data thinning operation>
In the following description, it is assumed that the lens parameters are d types, that is, the lens parameters are the first axis to the d-th axis. In the d-axis lens parameter, the number of measurement points is F _d .

As shown in FIG. 7, the chromatic aberration of magnification-image distortion correction apparatus 1, the correction data thinning means 3 counts the number of measurement points F ₁ for the first axis of the lens parameters (step S11).

Further, the lateral chromatic aberration / image distortion correction device 1 sets a correction data plane to be thinned out for the first axis of the lens parameter by the correction data thinning means 3 (step S12). That is, the correction data decimation means 3 sequentially decimates one correction data plane, two decimations, three decimations each time the process of step S12 is performed for the first axis of the lens parameter,... , F ₁ -2, for example, the number of correction data planes to be thinned is set. Here, when two or more correction data planes are thinned out, the correction data thinning means 3 also sets all combinations of two or more correction data planes to be thinned out from the entire first axis of lens parameters. Further, the correction data thinning means 3 also sets which correction data plane is thinned.

Further, the lateral chromatic aberration / image distortion correction apparatus 1 counts the number of measurement points F _d-1 for the d−1 axis of the lens parameter by the correction data thinning means 3 (step S13).

Further, the lateral chromatic aberration / image distortion correction device 1 sets a correction data plane to be thinned out about the d-1 axis of the lens parameter by the correction data thinning means 3 (step S14). That is, the correction data decimation means 3 sequentially decimates one correction data plane, two decimations, three decimations each time the process of step S14 is performed for the d-1 axis of the lens parameter. ..., F _d−1 −2 The number of correction data planes to be thinned out is set as in the case of thinning out. Here, when two or more correction data planes are thinned out, the correction data thinning means 3 also sets all combinations of two or more correction data planes to be thinned out from the entire lens parameter d-1 axis. Further, the correction data thinning means 3 also sets which correction data plane is thinned.

Further, lateral chromatic aberration, image distortion correction apparatus 1, the correction data thinning means 3 counts the number of measurement points F _d About d-axis of the lens parameters (step S15).

Further, the magnification chromatic aberration / image distortion correction device 1 sets a correction data plane for the d-th axis of the lens parameter by the correction data thinning means 3 (step S16). That is, the correction data thinning unit 3 sequentially thins out one correction data plane, thins out two, thins out three, and so on each time the process of step S16 is performed for the d-th axis of the lens parameter. , F _d -2 is set, the number of correction data planes to be thinned is set. Here, when two or more correction data planes are thinned out, the correction data thinning means 3 also sets all combinations of two or more correction data planes to be thinned out from the entire lens parameter d-axis. Further, the correction data thinning means 3 also sets which correction data plane is thinned.

  Further, the magnification chromatic aberration / image distortion correction apparatus 1 selects a correction data plane to be a thinning candidate from the correction data planes set in step S12, step S14 and step S16 by the correction data thinning means 3 (step S17). Here, when one correction data plane is thinned out, the correction data thinning means 3 performs the lens parameter in the order of the d-th axis, d-1th axis,..., First axis every time the process of step S17 is performed. Then, the value of m is incremented. Further, when two or more correction data planes are thinned out, the correction data thinning means 3 performs steps S12, S14 and step in order of the lens parameter d-axis, d-1 axis,. A selection is made for each combination of correction data planes set in S16.

  Further, the lateral chromatic aberration / image distortion correction device 1 calculates the evaluation value candidate Ev_temp of the correction data plane as a thinning candidate by the correction data thinning means 3 (step S18). That is, the correction data thinning unit 3 interpolates correction data planes that are thinning candidates by the same interpolation method as the correction amount calculation unit 8 to obtain post-interpolation correction data planes. Then, the correction data thinning means 3 calculates the aberration amount difference between the correction data plane and the thinning candidate correction data plane for all pixels. Further, the correction data thinning-out means 3 calculates the evaluation value candidate Ev_temp from among all the pixels that has the maximum difference.

  Further, the magnification chromatic aberration / image distortion correction device 1 has evaluated all the correction data planes by the correction data thinning means 3, that is, has calculated the evaluation value candidate Ev_temp for all the correction data planes. Is determined (step S19). Here, when all the evaluations are made (Yes in step S19), the lateral chromatic aberration / image distortion correction apparatus 1 proceeds to the process of step S20. On the other hand, when not evaluating all (No in step S19), the lateral chromatic aberration / image distortion correction apparatus 1 returns to the process of step S17.

Further, the lateral chromatic aberration / image distortion correction device 1 uses the correction data thinning means 3 to evaluate the evaluation value EV, the total number of planes = F ₁ * F ₂ * _... * F _d, and the reduced plane number DP (F _1, F2 _{,... Fd} ) are calculated (step S20). At this time, the correction data thinning-out means 3 calculates the minimum value of the evaluation value candidate Ev_temp as the evaluation value EV. Further, the correction data thinning-out means 3 may calculate the average value of the evaluation value candidates Ev_temp as the evaluation value EV.

Further, lateral chromatic aberration, image distortion correction apparatus 1, the correction data thinning means 3, the measurement points F _d to subtract "1" (step S21).

Further, lateral chromatic aberration, image distortion correction apparatus 1 is checked by the correction data thinning means 3, the measurement points F _d is whether it is smaller than "2" (step S22). Here, if the number of measurement points _{F d} is less than "2" (Yes in step S22), and lateral chromatic aberration, image distortion correction apparatus 1, the process proceeds to step S23. On the other hand, if the number of measurement points _{F d} is not less than "2" (No in step S22), and lateral chromatic aberration, image distortion correction apparatus 1 returns to the process of step S16.

Further, the magnification chromatic aberration / image distortion correction apparatus 1 subtracts “1” from the number of measurement points F _d−1 by the correction data thinning-out means 3 (step S23).

Further, the magnification chromatic aberration / image distortion correction device 1 determines whether or not the number of measurement points F _d-1 is less than “2” by the correction data thinning means 3 (step S24). Here, when the number of measurement points F _d-1 is less than “2” (Yes in step S24), the magnification chromatic aberration / image distortion correction apparatus 1 proceeds to the process of step S25. On the other hand, when the number of measurement points F _d-1 is not less than “2” (No in step S24), the lateral chromatic aberration / image distortion correction apparatus 1 returns to the process of step S14.

The magnification chromatic aberration / image distortion correction apparatus 1 subtracts “1” from the number of measurement points F ₁ by the correction data thinning-out means 3 (step S25).

Further, lateral chromatic aberration, image distortion correction apparatus 1, the correction data thinning means 3 determines the number of measurement points F ₁ is whether it is smaller than "2" (step S26). Here, if the number of measurement points _{F 1} is smaller than "2" (Yes in step S26), the chromatic aberration of magnification-image distortion correction apparatus 1, the process proceeds to step S23. On the other hand, if the number of measurement points _{F 1} is not less than "2" (No in step S26), the chromatic aberration of magnification-image distortion correction apparatus 1 returns to the process of step S12.

Further, the magnification chromatic aberration / image distortion correction device 1 outputs the evaluation value EV, the reduced plane number DP, and the total number of planes by the correction data thinning means 3 (step S27).
Further, the correction data thinning-out means 3 calculates the number of memories based on, for example, a value obtained by dividing the storage capacity of the correction data storage means 6 by the capacity per correction data plane, and the total number of planes is equal to or less than the number of memories. This is thinned out (step S28). At this time, the correction data thinning-out means 3 calculates the storage number so that the evaluation value EV is not more than the allowable value of the maximum color misregistration amount and the number of correction data planes to be thinned out is the maximum, and this correction number EV or less. It may be thinned out so that

  As described above, the magnification chromatic aberration / image distortion correction apparatus 1 can thin out correction data by the correction data thinning means 3 even when the lens parameters are of a plurality of types. For example, as shown in FIG. 8, the lens parameter of the first axis (depth direction) is zoom, the lens parameter of the second axis (horizontal direction) is iris, and the lens parameter of the third axis (vertical direction) is focus. Even in this case, the lateral chromatic aberration / image distortion correction apparatus 1 can thin out the correction data.

<Image correction operation>
As shown in FIG. 9, the lateral chromatic aberration / image distortion correction apparatus 1 inputs a G signal, an R signal, and a B signal by a signal input unit 4. The magnification chromatic aberration / image distortion correction apparatus 1 inputs photographing lens parameters by the lens parameter input means 5 (signal / lens parameter input step: step S101).

  Further, the magnification chromatic aberration / image distortion correction apparatus 1 selects correction data stored in the correction data storage unit 6 based on the lens parameter input to the lens parameter input unit 5 by the correction data selection unit 7 (correction). Data selection step: Step S102). Here, the correction data selection means 7 selects the correction data of the lens parameter that sandwiches the input lens parameter value among the correction data stored in the correction data storage means 6.

  Further, the magnification chromatic aberration / image distortion correction apparatus 1 calculates the correction amounts of the pixels of the G signal, the R signal, and the B signal by the correction amount calculation unit 8 based on the correction data selected by the correction data selection unit 7 ( Correction amount calculating step: Step S103). Here, the correction amount calculation means 8 calculates the correction amount in the input lens parameter from the aberration amount corresponding to the pixel from each of the plurality of correction data by linear interpolation or spline interpolation.

  Further, the lateral chromatic aberration / image distortion correction device 1 is configured to output the G signal, the B signal, and the R signal input to the lens parameter input unit 5 based on the correction amount calculated by the correction amount calculation unit 8 by the aberration correction unit 9. The color shift and image distortion due to lateral chromatic aberration are corrected and stored in the signal storage means 10 (aberration correction step: step S104). Here, the lateral chromatic aberration / image distortion correction apparatus 1 extracts an integer component from the correction amount calculated by the correction amount calculation means 8 by the shift processing unit 71 and performs a shift process based on the integer component. Further, the lateral chromatic aberration / image distortion correction apparatus 1 extracts a decimal component from the correction amount calculated by the correction amount calculation means 8 by the two-dimensional filter unit 72, and the pixel of the pixel shifted by the integer component of the correction amount Calculate the value. Then, the magnification chromatic aberration / image distortion correction apparatus 1 writes the pixel value to the address indicating the pixel subjected to the shift process in the signal storage unit 10 by the two-dimensional filter unit 72. As described above, the lateral chromatic aberration / image distortion correction apparatus 1 can correct the pixel value for each pixel for the G signal, the B signal, and the R signal.

  Then, returning to step S101, the magnification chromatic aberration / image distortion correcting apparatus 1 inputs the pixel value of the pixel next to the G signal, the B signal, and the B signal, and inputs the lens parameter corresponding to this pixel in step S101. Repeat the process.

  Further, the lateral chromatic aberration / image distortion correction apparatus 1 reads out the pixel values of the G signal, B signal, and R signal stored in the signal storage means 10 by the signal output means 11 to correct the corrected G signal, the corrected R signal, and the correction. Output as a B signal (signal output step, not shown).

  As described above, the magnification chromatic aberration / image distortion correction apparatus 1 according to the present embodiment stores correction data in consideration of the color shift and image distortion due to the magnification chromatic aberration in the correction data storage unit 6, and thus the image captured by the camera. The color shift due to the chromatic aberration of magnification and the image distortion can be corrected. The lateral chromatic aberration / image distortion correction apparatus 1 stores an appropriate number of correction data in the correction data storage unit 6, so that the number of correction data stored in the correction data storage unit 6 is neither too much nor too small. As a result, the magnification chromatic aberration / image distortion correction apparatus 1 makes the correction amount calculated using the correction data close to the actual shift amount of the video image, enabling accurate correction, and correcting the data stored in the correction data storage means 6. The storage capacity can be saved and the configuration of the magnification chromatic aberration / image distortion correction apparatus 1 can be simplified.

  Note that the magnification chromatic aberration / image distortion correction apparatus 1 can also implement each unit as a function program in a general computer, and combine the function programs to operate as a magnification chromatic aberration / image distortion correction program. It is also possible.

In addition, although it demonstrated as correcting both the color shift and image distortion by magnification chromatic aberration, the magnification chromatic aberration and image distortion correction apparatus 1 corrects only one of the color shift and image distortion by magnification chromatic aberration. You can also.
For example, a process of matching green (first primary color) with red and blue (second primary color and third primary color) is considered. In this case, the correction data is set to “0” (that is, no correction) for green (first primary color), and for red and blue (second primary color and third primary color), aberration with respect to green (first primary color) light. It is stored in the correction data storage means 6 (6a, 6b, 6c) as an amount. Thereby, the lateral chromatic aberration / image distortion correction apparatus 1 can correct only the color shift caused by the lateral chromatic aberration. A single-plate camera that does not include an imaging element for each primary color is less affected by image distortion because the imaging element does not rotate and tilt. Therefore, the lateral chromatic aberration / image distortion correction apparatus 1 is preferably applied to a single-plate camera.
Further, for example, the magnification chromatic aberration / image distortion correction apparatus 1 stores correction data in consideration of only image distortion in the correction data storage means 6 (6a, 6b, 6c), thereby correcting only image distortion. You can also.

DESCRIPTION OF SYMBOLS 1 Correction | amendment chromatic aberration / image distortion correction device 2 Correction data generation means 3 Correction data thinning means 4 Signal input means 5 Lens parameter input means 6 Correction data storage means 7 Correction data selection means 8 Correction amount calculation means 9 Aberration correction means (correction means)
10 signal storage means 11 signal output means

Claims (4)

The lens parameter value of the camera that shoots the image is associated with the aberration amount indicating at least one of the image position displacement amount and the image distortion amount due to the chromatic aberration of magnification for each pixel of the image in the lens parameter. The correction chromatic aberration / image distortion correction device for correcting at least one of color shift caused by the chromatic aberration of magnification and distortion of the image using the correction data,
Correction data storage means for storing the correction data of the first primary color, the second primary color and the third primary color which are the three primary colors;
A plurality of the correction data are input, and the number of correction data stored is calculated based on the storage capacity of the correction data storage means or a predetermined evaluation value, and the input correction data is thinned below the storage number. Correction data thinning means to be written to the correction data storage means from
Signal input means for inputting a first primary color signal, a second primary color signal, and a third primary color signal indicating the brightness of the first primary color, the second primary color, and the third primary color for each pixel;
Lens parameters for inputting photographing lens parameters corresponding to the pixels of the first primary color signal, the second primary color signal, and the third primary color signal input from the signal input means, which are lens parameters when the video is captured. Input means;
Correction data selection means for selecting correction data for two lens parameters sandwiching the value of the photographing lens parameter stored in the correction data storage means based on the photographing lens parameter input from the lens parameter input means;
Based on the correction data selected by the correction data selecting means, a correction amount for interpolating and calculating a correction amount for correcting at least one of the color shift due to the magnification chromatic aberration and the distortion of the image in the photographing lens parameter A calculation means;
Correction means for correcting a pixel value indicating the brightness of each pixel of the first primary color signal, the second primary color signal, and the third primary color signal based on the correction amount calculated by the correction amount calculation means;
Signal output means for outputting the first primary color signal, the second primary color signal, and the third primary color signal whose pixel values are corrected by the correction means;
A chromatic aberration magnification / image distortion correction apparatus comprising:   2. The magnification according to claim 1, wherein the correction data thinning-out means calculates the storage number based on a value obtained by dividing the storage capacity of the correction data storage means by the capacity per one piece of the correction data. Chromatic aberration / image distortion correction device.   The correction data thinning means calculates the aberration amount difference between the lens parameter value and the correction data of the thinning candidate for each pixel and the post-interpolation correction data obtained by interpolating the correction data, and for each lens parameter value. The maximum value or average value of the differences is calculated as the evaluation value, the evaluation value is equal to or less than a preset threshold value, and the maximum number of correction data of the thinning candidates is calculated as the storage number. The lateral chromatic aberration / image distortion correcting apparatus according to claim 1. The lens parameter value of the camera that shoots the image is associated with the aberration amount indicating at least one of the image position displacement amount and the image distortion amount due to the chromatic aberration of magnification for each pixel of the image in the lens parameter. The correction data of the first primary color, the second primary color, and the third primary color, which are the three primary colors, are stored in order to correct at least one of the color shift due to the chromatic aberration of magnification and the distortion of the image using the correction data. A computer comprising correction data storage means;
A plurality of the correction data are input, and the number of correction data stored is calculated based on the storage capacity of the correction data storage means or a predetermined evaluation value, and the input correction data is thinned below the storage number. Correction data thinning means for writing to the correction data storage means from
A signal input means for inputting a first primary color signal, a second primary color signal, and a third primary color signal indicating brightness of the first primary color, the second primary color, and the third primary color for each pixel;
Lens parameters for inputting photographing lens parameters corresponding to the pixels of the first primary color signal, the second primary color signal, and the third primary color signal input from the signal input means, which are lens parameters when the video is captured. Input means,
Correction data selection means for selecting correction data for two lens parameters sandwiching the value of the photographing lens parameter stored in the correction data storage means based on the photographing lens parameter input from the lens parameter input means;
Based on the correction data selected by the correction data selecting means, a correction amount for interpolating and calculating a correction amount for correcting at least one of the color shift due to the magnification chromatic aberration and the distortion of the image in the photographing lens parameter Calculation means,
Correction means for correcting a pixel value indicating the brightness of each pixel of the first primary color signal, the second primary color signal, and the third primary color signal based on the correction amount calculated by the correction amount calculation means;
Signal output means for outputting the first primary color signal, the second primary color signal and the third primary color signal whose pixel values have been corrected by the correction means;
A chromatic aberration / image distortion correction program for magnification characterized by functioning as

Images