JP5602467B2 - Center position determination method and center position determination apparatus - Google Patents

Description

  The present invention relates to a center position determination method and a center position determination device that detect the center position of a circular object, and in particular, even when dirt or a lack of an outer peripheral portion is captured in an image of a captured circular object. The present invention relates to a center position determination method and a center position determination apparatus that can detect the center position of a circular object with high accuracy.

  Conventionally, there has been known a center position determination method for determining a center position of an image of a circular object by using a histogram or the like obtained by capturing an image of a circular object with a camera and adding each pixel value of the captured image in a predetermined direction. ing.

  For example, in Patent Document 1, an image is picked up by irradiating light to a coin to be transported, and based on the brightness of each pixel constituting the image of the picked up coin (hereinafter referred to as “luminance”). A center position determination method for determining the center position of a coin is disclosed.

  Patent Document 2 discloses a coin that determines the center position of a coin by the same method as Patent Document 1, and cuts out an image of a pattern portion indicating the diameter of the coin and the characteristics of the coin based on the determined center position. An authenticity determination device is disclosed.

  In such a coin authenticity determination device, the denomination is determined from the detected diameter, and the authenticity of the coin is determined by comparing the cut pattern portion with a template stored in advance in the denomination.

JP-A-9-259320 Japanese Patent No. 3718613

  However, the conventional center position determination method has a problem that the accuracy of determining the center position is low. Specifically, when a coin floats for some reason during conveyance, the outer periphery of the imaged coin may become unclear due to dust adhering or the like.

  In this case, in the center position determination method of Patent Document 1, for example, the internal pattern portion is erroneously determined as the outer peripheral portion of the coin. Moreover, in the center position determination method of patent document 1, when dust has adhered to the outer peripheral part of a coin, the outer peripheral part including dust will be detected.

  As described above, in the center position determination method of Patent Document 1, a position different from the center position of the original coin is determined as the center position when the outer peripheral portion is erroneously detected. For this reason, the coin authenticity determination apparatus of Patent Document 2 using the center position determination method of Patent Document 1 has a problem that the accuracy of denomination determination and authenticity determination is low.

  From these things, the center position determination method and center that can detect the center position of the circular object with high accuracy even when the image of the captured circular object has an image of dirt or lack of the outer peripheral portion. How to realize the position determining device is a big problem.

  The present invention has been made to solve the above-described problems caused by the prior art, and is highly accurate even when the image of a circular object is imaged of dirt or lack of the outer peripheral portion. Another object of the present invention is to provide a center position determination method and a center position determination apparatus that can detect the center position of a circular object.

In order to solve the above-described problems and achieve the object, the present invention is a center position determination method applied to a center position determination device that detects the center position of a circular object, and an image for acquiring an image of the circular object An acquisition step, a temporary center setting step for setting a temporary center of the circular object based on the characteristics of the image acquired by the image acquisition step, and the image for the temporary center set by the temporary center setting step. A partial image surrounded by two concentric circles set so as to include the outer periphery of the circular object having the largest and smallest diameters among the plurality of kinds of circular objects whose center position is to be determined in the image acquired by the acquisition process is subjected to polar coordinate conversion. a polar coordinate image acquisition step of acquiring an image, and an outer peripheral detection step of detecting a trajectory of the outer periphery of the circular object based on images acquired by the polar coordinate image acquisition step, wherein Characterized in that it includes a central correction step of correcting a center of the tentative of the circular object to the true center position based on the trajectory of the outer peripheral portion which is detected by the peripheral detection step.
In the present invention, the provisional center setting step uses a horizontal and vertical luminance histogram of the image of the circular object.

The present invention, in the above invention, before Kigaishu detecting step, and detects the outer peripheral portion on the basis of the partial image that is polar coordinate transformed by the polar coordinate image acquisition process.
Also, the present invention, in the above invention, the image acquisition process is characterized by dealing a predetermined region of the obtained circular was outside yelling reflected image as invalid data.

In the present invention, the outer periphery detection step further includes an outer periphery correction step of correcting the trajectory by smoothing the detected trajectory, and the center correction step includes the outer periphery correction step. and correcting the center of the tentative of the circular object to the true center position based on the corrected the orbit by.
Further, in the present invention according to the above invention, in the center correction step, the distance between the two points where the phase difference on the orbit of the outer peripheral portion is 180 degrees is calculated for a plurality of locations on the outer peripheral portion. Then, the provisional diameter length of the circular object is obtained based on the calculated distance.
Further, the present invention is characterized in that, in the above-mentioned invention, the center correction step further obtains a reference radial length from an average of a predetermined number satisfying a predetermined condition among the plurality of temporary radial lengths.
Further, according to the present invention, in the above-described invention, in the outer periphery correction step, the two radii are calculated by comparing two radii calculated for the outer peripheral portions of both end portions of the diameter and half of the reference radial length. Each is corrected.

Further, the present invention provides a periodic function representing a distance from a predetermined point on the circumference whose center is the true center position and whose radius is half the reference radial length to the temporary center in the above invention. The amount of deviation between the true center position and the temporary center is calculated by substituting the value of the point on the trajectory, and the temporary center is converted to the true center position using the calculated amount of deviation. It is characterized by correcting to.

According to the present invention, there is provided a center position determination method applied to a center position determination device that detects a center position of a circular object, wherein the center position determination device acquires an image of the circular object, and the acquired image has a feature. The center position determination device sets the temporary center of the circular object based on the image , and the image acquired for the set temporary center is the circular object having the largest and smallest diameters among a plurality of types of circular objects whose center position is to be determined. The center position determination device acquires an image obtained by polar-coordinate conversion of a partial image surrounded by two concentric circles set so as to include the outer periphery of the center, and the center position determination device determines the trajectory of the outer periphery of the circular object based on the acquired image. Since the center position determination device corrects the temporary center of the circular object to the true center position based on the detected outer track of the outer periphery , the image of the circular object is The missing part is imaged Even when there is an effect that it is possible to detect the center position of the circular object with high accuracy.

  Further, according to the present invention, two concentric circles centered on the temporary center are set so as to include at least the outer peripheral portion, and the center position determining device performs polar coordinate conversion on the partial image surrounded by the two concentric circles set. Since the center position determination device detects the outer periphery based on the partial image subjected to polar coordinate conversion, it is possible to reduce the storage capacity and processing time required for the outer periphery detection processing.

  According to the present invention, the center position determining device corrects the detected trajectory by smoothing the detected trajectory, and the center position of the temporary center of the circular object is changed to the true center position based on the corrected trajectory. Since the determination device corrects, it is possible to detect the center position of the circular object with high accuracy even when the image of the captured circular object has an image of dirt or lack of the outer peripheral portion. There is an effect.

  In addition, according to the present invention, the distance between the two points where the phase difference on the orbit of the outer peripheral portion is 180 degrees is calculated for a plurality of locations on the outer peripheral portion, and the circular object is calculated based on the calculated distance. Since the center position determining device determines the temporary diameter, there is an effect that the temporary center of the circular object can be corrected with high accuracy.

  Further, according to the present invention, a point on the orbit is substituted for a periodic function representing a distance from a predetermined point on the circumference whose center is a true center position and whose radius is a temporary diameter to the temporary center. Thus, the deviation amount between the true center position and the temporary center is calculated, and the center position determination device corrects the temporary center to the true center position using the calculated deviation amount. By performing the correction, the center position of the circular object can be detected with high accuracy.

FIG. 1 is a diagram showing an outline of a center position determination method according to the present invention. FIG. 2 is a block diagram illustrating the configuration of the coin processing device according to the present embodiment. FIG. 3 is a diagram for explaining provisional center setting. FIG. 4 is a diagram illustrating an example of cutting out a partial image. FIG. 5 is a diagram for explaining polar coordinate conversion processing of a partial image. FIG. 6 is a diagram illustrating an example of an outer peripheral portion detection method. FIG. 7 is a diagram illustrating an example of the outer peripheral portion correction method 1. FIG. 8 is a diagram illustrating an example of a reference diameter calculation method. FIG. 9 is a diagram illustrating an example of a temporary center correction method. FIG. 10 is a diagram illustrating an example of the outer peripheral portion correction method 2. FIG. 11 is a flowchart showing the center position determination processing procedure executed by the coin processing device. FIG. 12 is a flowchart showing an outer periphery / center position correction processing procedure executed by the coin processing device. FIG. 13 is a diagram illustrating a modification example of cutting out a partial image.

  Exemplary embodiments of a center position determination method and a center position determination apparatus according to the present invention will be described below in detail with reference to the accompanying drawings. First, prior to detailed description of the embodiment, an outline of the center position determination method according to the present invention will be described with reference to FIG. Hereinafter, the case where the circular object is a coin will be described.

  FIG. 1 is a diagram showing an outline of a center position determination method according to the present invention. In the following, first, an outline of the conventional center position determination method will be described using (A) of the figure, and an outline of the center position determination method according to the present invention will be explained using (B) of the figure. 1A and 1B are views of a coin image as viewed from above. Hereinafter, description will be made by appropriately using coordinate axes as shown in the upper right of the figure.

  First, as shown in FIG. 1A, in the conventional center position determination method, an image of a coin is captured, and the brightness (hereinafter referred to as “brightness”) of each pixel constituting the captured image of the coin. The outer periphery of the coin is detected using a histogram obtained by adding (described) in a predetermined direction.

  Specifically, a histogram (see the lower part of FIG. 1A) in which the luminance of each pixel is added in the Y-axis direction is created, and the position where the luminance of the histogram exceeds the threshold value from the positive direction side of the X-axis to the negative direction Is the outer periphery A of the coin. Further, the position where the brightness of the histogram exceeds the threshold value in the positive direction from the negative direction side of the X axis is defined as the outer peripheral portion B of the coin.

  And in the conventional center position determination method, the midpoint of the outer peripheral part A and the outer peripheral part B is made into the X coordinate of the center position of a coin. As for the Y coordinate, the Y coordinate of the center position of the coin is obtained based on a histogram obtained by adding the luminance of each pixel in the X-axis direction.

  However, in the conventional center position determination method, as shown in FIG. 1A, when dust is attached to the edge of the coin, the brightness of the imaged dust portion is also added to the histogram, and X From the positive direction side of the shaft, the outer peripheral portion A outside the actual outer peripheral portion is detected. Therefore, in the conventional center position determination method, a position different from the actual center position may be set as the center position.

  Moreover, in the coin authenticity determination device using the conventional center position determination method, the accuracy of the denomination determination and authenticity determination result performed based on the center position obtained by the center position determination method described above is also improved. It will be missing.

  Therefore, in the center position determination method according to the present invention, a captured image is converted from a format represented by polar coordinates onto a two-dimensional orthogonal coordinate plane, and all outer peripheral portions are detected from the image thus converted. Then, the temporary center is corrected to the true center position based on the detected orbit of the outer peripheral portion.

  Specifically, the center position determination method according to the present invention first calculates a temporary center as shown in FIG. Here, as a temporary center calculation method, a method based on the histogram of the luminance of each pixel constituting the captured coin image can be used as in the conventional case.

  Then, as shown by the dotted line (B-1) in FIG. 1, the center position determination method according to the present invention cuts out an image of a coin in a portion surrounded by two concentric circles centered on a temporary center. Then, polar coordinate conversion is performed on the cut out partial image (see (B-2) in FIG. 1). Note that the diameters of the two concentric circles are adjusted so that the cut out partial image includes at least the outer peripheral portion of the coin. Further, (B-2) in FIG. 1 shows an arrow indicating a temporary center direction together with an image subjected to polar coordinate conversion.

  The center position determination method according to the present invention detects an outer peripheral portion from an image converted from a format represented by polar coordinates onto a two-dimensional orthogonal coordinate plane. For example, when the luminance of each pixel is sequentially searched from the negative direction of the Y-axis on the orthogonal coordinate thus transformed to the positive direction and a predetermined threshold value is exceeded, this pixel is set as the outer peripheral portion.

  The outer periphery trajectory thus detected draws a straight line when the temporary center and the true center position coincide. On the other hand, when the temporary center is different from the true center position, as shown in FIG. 1 (B-2), the trajectory of the outer peripheral portion draws a curve represented by a periodic function such as a trigonometric function.

  Therefore, the center position determination method according to the present invention obtains a deviation between the temporary center and the true center position based on the outer periphery trajectory, and corrects the temporary center to the true center position. (See (B-3) in FIG. 1).

  Further, in the center position determination method according to the present invention, as shown in FIG. 1B-2, when the outer peripheral portion is erroneously detected due to dust adhesion or the like, the outer peripheral portion trajectory is smoothed. A process for correcting the outer peripheral portion is also performed, which will be described later. In FIG. 1B-2, as an example of erroneous detection of the outer peripheral portion, a case where dust is imaged on the outer peripheral portion is given. However, when a coin floats or the outer peripheral portion is chipped. In some cases, similar correction processing is performed.

  As described above, in the center position determination method according to the present invention, the center position of the circular object is detected with high accuracy even when the image of the captured circular object includes an image of dirt or lack of the outer peripheral portion. be able to.

  Below, the Example about the coin processing apparatus to which the center position determination method demonstrated using FIG. 1 is applied is described in detail. First, the structure of the coin processing apparatus 10 according to the present embodiment will be described with reference to FIG.

  FIG. 2 is a block diagram illustrating the configuration of the coin processing device according to the present embodiment. In FIG. 2, only constituent elements necessary for explaining the characteristic points of the coin processing apparatus 10 are extracted and shown.

  The coin processing device 10 includes an imaging unit 11, a storage unit 13, and a control unit 14. The storage unit 13 stores image information 13a, center position information 13b, and outer periphery information 13c, and the control unit 14 includes an image acquisition unit 14a, a temporary center setting unit 14b, and an outer periphery cutout unit 14c. The outer periphery detection unit 14d, the outer periphery correction unit 14e, the center correction unit 14f, and the output unit 14g are further provided.

  The imaging unit 11 is a light irradiation device that irradiates a coin with a plurality of types of light such as green light and visible light and infrared light, and a CCD that receives reflected light reflected from a bill by light emitted from the light irradiation device. Charge Coupled Device) camera. And the imaging part 11 images the image of a coin, and passes the imaged image data to the image acquisition part 14a.

  The storage unit 13 is a storage device such as a nonvolatile memory or a hard disk drive, and stores image information 13a, center position information 13b, and outer periphery information 13c. The image information 13a is image data of a coin imaged by the imaging unit 11.

  The center position information 13b is center position information corrected by the center correction unit 14f. The outer periphery information 13c is data related to the outer periphery such as the track of the outer periphery of the coin and the diameter of the coin detected by the outer periphery detection unit 14d.

  The control unit 14 is a control unit that performs overall control of the coin processing apparatus 10. The image acquisition unit 14 a is a processing unit that performs processing for acquiring image data of a coin imaged by the imaging unit 11 and passing it to the image information 13 a of the storage unit 13.

  The temporary center setting unit 14b sets a temporary center based on the coin image data acquired by the image acquisition unit 14a. The details of the method for setting the temporary center will be described later.

  The outer periphery cutout unit 14c is a processing unit that performs a process of cutting out a partial image including the outer periphery from the image information 13a. In addition, the outer peripheral cutout unit 14c also performs a process of performing polar coordinate conversion on the cutout partial image based on the temporary center set by the temporary center setting unit 14b or the center position information 13b. The details of the outer peripheral cutting method will be described later.

  The outer periphery detection unit 14d searches for a pixel that has been subjected to polar coordinate conversion by the outer periphery cutout unit 14c in a predetermined direction, and determines that the pixel is an outer periphery when the luminance of the pixel exceeds a predetermined threshold for two consecutive pixels. It is a processing unit that performs processing for detection. The outer periphery detection unit 14d also performs a process of passing the detected information on the outer periphery to the outer periphery information 13c in the storage unit 13. The details of the outer periphery detection method will be described later.

  The outer periphery correction unit 14e is a processing unit that performs a process of correcting the outer periphery of the outer periphery to be smoothed when the outer periphery detected by the outer periphery detection unit 14d is a foreign matter portion such as dust. Here, when the temporary center is different from the true center position, the trajectory of the outer peripheral portion is represented by a predetermined periodic function.

  Therefore, in the coin processing apparatus 10, a curve that draws a periodic function is prepared in advance as a template, and the outer periphery correction unit 14e compares the outer track with the template to correct a foreign matter portion such as dust. Good. Further, a foreign matter portion such as dust may be corrected by comparing a value obtained by integrating a predetermined periodic function with a value at the outer peripheral portion. The details of the outer peripheral correction method will be described later.

  The center correction unit 14f is a processing unit that performs a process of correcting the temporary center to the true center position. Specifically, the outer track is represented by a predetermined periodic function, and the cycle between the maximum value and the minimum value of the track is 180 °. Therefore, the center correction unit 14 f corrects the temporary center to the true center position based on the outer periphery orbit and passes it to the center position information 13 b of the storage unit 13.

  Further, the center correction unit 14f passes an instruction to cut out the outer periphery to the outer periphery cutout unit 14c based on the corrected center position. The details of the center position correction method will be described later.

  The output unit 14g cuts out the outer peripheral portion from the image information 13a again based on the center position information 13b, and a coin identification processing unit (not shown) is obtained as a result of detecting the outer peripheral portion again by the outer peripheral detecting portion 14d. It is a processing part which performs the process passed to.

  The denomination identification processing unit identifies the denomination of the coin based on the outer peripheral portion detected based on the center position corrected by the center correction unit 14f, the image information 13a, and the like. Moreover, the output part 14g is good also as outputting the result of having detected the outer peripheral part again to the authenticity determination process part which is not shown in figure which determines the authenticity of a coin.

  Next, FIG. 3 to FIG. 3 show the details of the provisional center setting method, outer periphery cutting method, outer periphery detection method, outer periphery correction method, and center position correction method executed by the coin processing device 10. 10 will be used for explanation.

  First, details of a method for setting a temporary center executed by the coin processing apparatus 10 will be described with reference to FIG. FIG. 3 is a diagram for explaining provisional center setting. Here, the case where the image of the coin conveyed on the conveyance path with which the coin processing apparatus 10 is equipped is imaged with the imaging part 11 is demonstrated. As shown in FIG. 3, suppose that a coin is conveyed in the conveyance direction shown by the arrow.

  First, as shown in the lower right of FIG. 3, the temporary center setting unit 14 b creates a horizontal histogram in which the luminance of each pixel constituting the captured coin image is added in a direction parallel to the transport direction. Then, the temporary center setting unit 14b searches from the left and right ends of the horizontal histogram shown in FIG. 3, and when the value of the horizontal histogram exceeds a predetermined threshold (here, threshold A), the temporary center setting unit 14b Part.

  In this way, the outer periphery detected when searching from the right to the left is the right outer periphery, and the outer periphery detected when searching from the left to the right is the left outer periphery. These are indicated by “triangles”.

  Next, the temporary center setting unit 14b creates a vertical histogram in which the luminance of each pixel constituting the captured coin image is added in a direction perpendicular to the transport direction. Here, since coins are continuously transported on the transport path, another coin may be transported by being overlapped with the front end portion or the end portion in the coin transport direction.

  An accurate outer periphery cannot be detected with the luminance of the pixels including such overlapping portions. Therefore, in order to exclude the value of the overlapping portion of coins, it is necessary to detect the outer peripheral portion with a width smaller than the maximum width of the end portion from the front end portion in the transport direction. In addition, since the number of pixels to be added in the portion for detecting the outer peripheral portion is larger than that in the case of detecting the outer peripheral portion from the horizontal histogram, the outer peripheral portion is detected by increasing the threshold value.

  Thereby, the temporary center setting unit 14b searches from the upper and lower ends of the vertical histogram shown in FIG. 3 and if the value of the vertical histogram exceeds a threshold value (here, the threshold value B) larger than the threshold value A, such a pixel is determined. The upper and lower outer peripheries are used. In FIG. 3, each is indicated by a “circle”.

  The temporary center setting unit 14b uses the midpoint between the right outer peripheral portion and the left outer peripheral portion as the coordinates of the axis perpendicular to the temporary center conveyance direction, and sets the midpoint between the upper outer peripheral portion and the lower outer peripheral portion as the temporary center. Set the coordinates of the axis parallel to the transport direction.

  Next, the details of the outer peripheral cutting method performed by the coin processing device 10 will be described with reference to FIGS. 4 and 5. FIG. 4 is a diagram illustrating an example of cutting out a partial image, and FIG. 5 is a diagram illustrating a polar coordinate conversion process of the partial image.

  First, the partial image cut-out process will be described. The outer peripheral cutout unit 14c is a partial image that is surrounded by two concentric circles centered on the temporary center set by the temporary center setting unit 14b from the image information 13a captured by the imaging unit 11 and includes the outer peripheral part. cut.

  For example, as shown in FIG. 4A, the cut-out partial image includes the outer periphery of the largest 500-yen coin among the denominations, and as shown in FIG. The partial image to be included includes the smallest outer peripheral portion of a 1-yen coin among the denominations. In this way, the radius of the concentric circle is set so that an image including any outer peripheral portion of coins having different sizes depending on the denominations can be cut out.

  Here, as shown by the dotted line in FIG. 4A, the outer periphery cutout portion 14c has a provisional center at the center of the circle and a circle with a radius of 50 pixels and a radius of 120 pixels (50 pixels + 70 pixels = 120). A partial image surrounded by a circle of pixels is cut out.

  Although the partial image surrounded by a circle with a radius of 50 pixels and a circle with a radius of 120 pixels was cut out, the maximum and minimum coin sizes differ depending on the currency. It is good also as changing the radius of a concentric circle.

  Next, polar coordinate conversion processing of the partial image will be described with reference to FIG. 5A shows a partial image to be cut out, and FIGS. 5B and 5C show images obtained by polar coordinate conversion. Hereinafter, description will be made using coordinate axes as shown on the right side of FIG. 5 as appropriate. I will do it.

  The outer periphery cutout unit 14c converts the cutout partial image (see FIG. 5A) from a format expressed in polar coordinates based on a temporary center onto a two-dimensional orthogonal coordinate plane (hereinafter, “polar coordinate conversion”). (Refer to FIG. 5B). Specifically, it is assumed that the X-axis of the orthogonal coordinates from the polar coordinates represented by a predetermined radius and the argument θ corresponds to the argument θ, and the Y-axis corresponds to the radius and is converted.

  Here, as shown in FIG. 5B, the negative direction of the Y axis is the center side of the circle, and from the 50th pixel to the 120th pixel on the same diameter from the negative direction of the Y axis to the positive direction. Image data is converted, and pixels constituting the circumference of the same radius are converted in the X-axis direction. Also, assuming that the number of pixels constituting the circumference of a predetermined radius is 512 pixels, the declination is converted at an interval of 0.703 degrees (360 ÷ 512 = 0.703). Note that the number of pixels varies depending on the accuracy of the imaging unit 11 and the like.

  Here, if all the pixel values converted into an array of 512 pixels in the X-axis direction and 71 pixels in the Y-axis direction are used, it takes time to correct the center position performed by the coin processing device 10.

  Therefore, in order to shorten the correction processing time, the outer peripheral cutout unit 14c extracts pixels at predetermined intervals for the X axis and sets them as processing targets. Here, a case will be described in which pixel values constituting the circumference at positions equally divided into 32 in the X-axis direction are extracted.

  In this case, pixel values in the radial direction are extracted at intervals of 11.25 ° (360 ÷ 32 = 11.25). The black part of FIG. 5B indicates the pixel of the nth angle (11.25 * n °) to be extracted, where the negative end of the X axis is 0 degree.

  When extracting the pixel values for 32 equal parts, the outer peripheral cutout unit 14c corrects pixel value fluctuation (hereinafter referred to as “noise”) due to dust or the like for pixels arranged in the X-axis direction. The following processing is performed.

  The outer peripheral cutout portion 14c includes pixels having an angle of 11.25 * n °, two pixels adjacent in the X-axis positive direction and two pixels adjacent in the X-axis negative direction (see the hatched portion in FIG. 5B). A value obtained by adding pixel values for a total of five pixels is set as a pixel value to be extracted.

  As described above, the outer peripheral cutout unit 14c extracts pixel values expressed by an array of 32 pixels in the X-axis direction and 71 pixels in the Y-axis direction from the pixel values (512 pixels × 71 pixels) subjected to polar coordinate conversion. (Refer to FIG. 5C).

  Specifically, the image data subjected to polar coordinate conversion is EDGE [x] [y] (x = 0 to 511, y = 0 to 70), and the image data extracted by correcting noise is DATA [x]. [y] (x = 0 to 31, y = 0 to 70), the equation (1) is obtained when x = 0, and the equation (2) is obtained when x> 0.

  Next, details of the outer peripheral portion detection method executed by the coin processing device 10 will be described with reference to FIG. FIG. 6 is a diagram illustrating an example of an outer peripheral portion detection method.

  The outer periphery detection unit 14d performs the following processing for x = 0 to 31 for DATA [x] [y], which is image data extracted by the outer periphery cutout unit 14c and extracted at 11.25 ° intervals. .

  FIG. 6 is a graph showing the luminance of each image data in which y = 0 to 70 of the image data DATA [x] [y] for a predetermined x. Here, the horizontal axis represents the y value, and the vertical axis represents the luminance value. Therefore, the outer periphery detection unit 14d searches for the luminance from the positive direction to the negative direction on the horizontal axis, and determines that the outer periphery is detected when the luminance of the image data exceeds the threshold continuously at two locations.

  Here, for the predetermined x, the maximum luminance of each image data in which y = 0 to 70 of the image data DATA [x] [y] is calculated for a predetermined x, and the maximum luminance / 8 is set as the threshold. The threshold may be determined using a calculation method.

  For example, as illustrated in FIG. 6, the outer periphery detection unit 14d searches from y = 70 in the direction of y = 0, and when y = m + 1 does not exceed the threshold, and when y = m and m−1 exceed the threshold, DATA [x] [m−1] is determined to be the outer peripheral portion. The outer periphery detection unit 14d performs the above-described determination process for each of x = 0 to 31, and detects the outer periphery of the image data for 32 locations.

  Note that the outer periphery detection unit 14d determines that the image data has an outer periphery when the luminance of the image data exceeds a predetermined threshold value for two consecutive locations. It may be determined. Further, the outer periphery detection unit 14d determines that DATA [x] [m−1] illustrated in FIG. 6 is the outer periphery, but may determine DATA [x] [m] as the outer periphery. .

  Next, details of the outer peripheral portion correction method and the center position correction method executed by the coin processing apparatus 10 will be described with reference to FIGS. 7 to 10. FIG. 7 is a diagram illustrating an example of the outer peripheral portion correction method, FIG. 8 is a diagram illustrating an example of a reference diameter calculation method, and FIG. 9 is a diagram illustrating the outer peripheral portion trajectory. FIG. 10 is a diagram illustrating an example of the outer peripheral portion correction method 2.

  Here, the case where the temporary center set by the temporary center setting unit 14b is different from the true center position will be described with reference to FIGS. 7 to 9, and the case where the temporary center matches the true center position. This will be described with reference to FIG.

  First, as shown in FIG. 7A, when the temporary center set by the temporary center setting unit 14b is different from the true center position, the outer periphery detected by the outer periphery detection unit 14d is as shown in FIG. The trajectory is as shown in (B). In addition, when dust is captured in the coin image captured by the imaging unit 11, the outer periphery detection unit 14d may determine that the dust is the outer periphery. Here, the circles shown in FIG. 7B indicate the outer periphery determined by the outer periphery detection unit 14d.

  Therefore, the outer periphery correcting unit 14e obtains a reference diameter (hereinafter referred to as “reference diameter”), and smoothes and corrects the outer periphery orbit based on a predetermined periodic function represented by the reference diameter. (See FIG. 7C).

  In the center correction unit 14f, since the orbit of the outer periphery is represented by a predetermined periodic function, the center position is obtained by substituting the corrected value on the orbit of the outer periphery into the periodic function. In the coin processing device 10, when the center position is obtained, the outer peripheral portion is cut out again by the outer peripheral cut-out portion 14c, and the outer peripheral portion is detected by the outer peripheral detection portion 14d from the cut-out image data, as shown in FIG. The outer periphery drawn with a straight line is detected, and the diameter can be obtained with high accuracy based on the outer periphery drawn with a straight line.

  Here, a method of calculating the reference diameter will be described with reference to FIG. FIG. 8 is a diagram illustrating an example of a reference diameter calculation method. First, as shown in FIG. 8, the outer periphery correcting unit 14 e has an outer peripheral portion (hereinafter referred to as “diagonal”) having an angular difference of 180 ° with respect to a predetermined outer peripheral portion of the 32 outer peripheral portions detected by the outer peripheral detecting unit 14 d. ”(Referred to as“ diagonal diameter ”).

  And the outer periphery correction | amendment part 14e sorts the calculated | required 16 diagonal diameters by diameter length, calculates the average of four center lengths, and makes it a reference | standard diameter length. Here, the average of the central four is calculated, but a predetermined number in the center, for example, the average of the two central may be calculated and used as the reference diameter length.

  In this way, the difference between the reference radial length calculated by the outer peripheral correction unit 14e and the 16 diagonal radial lengths is calculated, respectively, and the trajectory of the outer peripheral portion is smoothed by the difference. Specifically, if the difference is greater than or equal to a predetermined value, the outer periphery correction unit 14e checks the radial length of the diagonal diameter, and if there is an abnormality in the length, estimates it from the preceding and subsequent data to obtain it. The outer circumference trajectory is corrected so as to be smooth.

  Furthermore, the outer periphery correction unit 14e corrects the two radii by comparing the two radii calculated for the outer periphery of the opposite end portions of the diagonal diameter with the reference diameter length / 2. Further, a foreign matter portion such as dust may be corrected by comparing a value obtained by integrating a predetermined periodic function with a value at the outer peripheral portion. In this way, by correcting the foreign matter portion included in the outer peripheral portion, it is possible to correct the center position described later with high accuracy.

  Then, the center correction unit 14f performs the following processing on the outer periphery corrected by the outer periphery correction unit 14e to correct the temporary center. The center correction unit 14f calculates the difference between the two radii calculated based on the outer peripheral portions of both end portions of the diagonal diameter and the reference radial length / 2 for the 16 diagonal diameters, and the absolute value of the difference (Hereinafter, this value is referred to as “amplitude sum”).

  Here, the center correction unit 14f determines that the correction of the center position is not required when the calculated maximum value of the amplitude sum is smaller than the predetermined value. Further, the center correction unit 14f corrects the temporary center when the calculated maximum value of the amplitude sum is larger than a predetermined value. Hereinafter, a temporary center correction method will be described with reference to FIG. FIG. 9 is a diagram illustrating an example of a temporary center correction method.

  As shown in FIG. 9A, the temporary center set by the temporary center setting unit 14b is defined as a point A (x, y), and two points that are the maximum of the amplitude sum calculated by the center correction unit 14f are obtained. Assume that the outer peripheral portions are point B and point C. Here, the longer length (line segment AB) from the temporary center A to the outer peripheral points B and C is rmax, and the shorter length (line AC) is rmin, and the line AB and the X axis are The argument formed is θmax, and the argument formed between the line segment AC and the X axis is θmin.

In this case, when the corrected center is (X, Y), the corrected center (X, Y) is as shown in equations (3) and (4).
X = x + (rmax−rmin) / 2 * cos (θmax) (3)
Y = y + (rmax−rmin) / 2 * sin (θmin) (4)

  Thus, the center correction unit 14f corrects the temporary center. Further, the graph shown in FIG. 9B is a graph showing the radius of the outer peripheral portion corresponding to the deflection angle calculated by the center correction unit 14f. The X axis is the declination, and the Y axis is the length of the calculated radius.

  Here, when the graph is moved in parallel in the X-axis direction from θmin to 360 °, as shown in FIG. 9C, the temporary center point A is corrected to the point A ′, and θmin is 0. ° and θmax are 180 °, and the line segment B′C ′ is a line segment parallel to the X axis. The temporary center may be corrected to (X, Y) by substituting the value corrected in this way into Equation (3) and Equation (4).

  As described above, the outer peripheral cutout unit 14c extracts pixels at predetermined intervals from the image data subjected to polar coordinate conversion, and corrects the center position based on the extracted image data. Then, the correction processing time can be shortened.

  So far, the case where the temporary center set by the temporary center setting unit 14b is different from the true center position has been described. Next, details of the outer peripheral correction method when the temporary center coincides with the true center position will be described with reference to FIG.

  FIG. 10 is a diagram illustrating an example of the outer peripheral portion correction method 2. First, as shown in FIG. 10A, when the temporary center set by the temporary center setting unit 14b coincides with the true center position, the outer periphery detected by the outer periphery detection unit 14d is as shown in FIG. As shown in (B) of FIG.

  In addition, when a part of the coin floats, the part that is not in close contact with the imaging surface is missing from the outer peripheral part in the image captured by the imaging unit 11 as shown in FIG. Will be imaged. However, the outer periphery correction unit 14e may detect the diameter where there is no foreign matter such as missing or dust.

  Next, a processing procedure executed by the above-described coin processing device 10 will be described with reference to FIGS. In FIG. 11, the center position determination process is described, and in FIG. 12, the outer periphery / center position correction process is described.

  FIG. 11 is a flowchart showing the center position determination processing procedure executed by the coin processing device 10. As shown in the figure, the image acquisition unit 14a acquires an image of a coin imaged by the imaging unit 11 (step S101).

  The temporary center setting unit 14b creates a histogram of the luminance of each pixel constituting the image based on the coin image acquired by the image acquisition unit 14a (step S102), and detects upper, lower, left, and right outer peripheral parts based on the histogram. (Step S103).

  Then, the temporary center setting unit 14b sets the midpoints of the upper and lower outer peripheral parts and the left and right outer peripheral parts as a temporary center in the coin image (step S104), and the outer peripheral part cutout part 14c is an outer peripheral part with respect to the temporary center. Is cut out (step S105).

  Thereafter, the outer periphery detection unit 14d detects the outer periphery based on the cut-out image (step S106), and determines whether or not the temporary center is deviated from the true center position by the detected outer track. Determination is made (step S107).

  If it is determined by the outer periphery detection unit 14d that the temporary center is shifted (step S107, Yes), the outer periphery / center position is corrected (step S108). The details of the outer periphery / center position correction processing will be described with reference to FIG.

  Subsequently, the outer periphery cutout unit 14c acquires the center position corrected by the outer periphery and center position correction processing (step S109), and cuts out an image including the outer periphery of the acquired center position (step S110). .

  And the outer periphery detection part 14d detects the outer periphery information 13c containing an outer periphery part and a diameter based on the cut-out image (step S111), passes the detected outer periphery information 13c to the output part 14g, and the coin processing apparatus 10 performs. The center position determination process ends.

  On the other hand, when it determines with the center position not having shifted | deviated by step S107 (step S107, No), the coin processing apparatus 10 transfers a process to step S111, and detects the outer periphery information 13c.

  Next, the outer periphery / center position correction processing in step S108 will be described with reference to FIG. FIG. 12 is a flowchart showing an outer periphery / center position correction processing procedure executed by the coin processing apparatus 10.

  As shown in FIG. 12, the outer periphery correcting unit 14e calculates diagonal lengths for the 32 outer peripheral portions detected by the outer periphery detecting unit 14d (step S201), and calculates the calculated 16 diagonal lengths. Are sorted by diameter length (step S202).

  And the outer periphery correction | amendment part 14e calculates 4 averages of the center of the sorted diagonal diameter length as a reference diameter length (step S203), and correct | amends an outer peripheral part based on a reference | standard diameter length (step S204), The center correction unit 14f corrects the temporary center to the true center position based on the corrected outer periphery (step S205).

  By the way, in the present embodiment, as described with reference to FIGS. 4 and 5, the outer peripheral cutout portion 14 c has two temporary centers centered on the temporary center set by the temporary center setting portion 14 b from the image information 13 a. A partial image surrounded by concentric circles was cut out.

  However, it is also possible to invalidate the image data of the area exceeding the width of the conveyance path through which the coin passes from the captured image so that the area outside the conveyance path is not detected as the outer peripheral portion. For example, the luminance of the pixels in such an area is set to 0 so that it is not detected as an outer peripheral portion. A specific method for extracting a partial image will be described with reference to FIG.

  FIG. 13 is a diagram illustrating a modification example of cutting out a partial image. As shown in the figure, the outer periphery of the coin is represented by a solid circle, and the outer circle of the concentric circle centered on the temporary center is represented by a broken line. In the following description, the coordinate axes as shown on the right side of FIG.

  When coins are transported in the Y-axis direction, the coin processing apparatus 10 sets a right line and a left line that are both ends of the transport path in advance. Then, the image data in the area located on the positive side of the X axis from the right line or on the negative side of the X axis from the left line as shown by the hatched portion in FIG. 13 is invalidated.

For example, when the x coordinate of a pixel (point A) as indicated by the “triangle” mark in FIG. 13 is less than the x coordinate of the left line or exceeds the x coordinate of the right line, an image of the pixel Invalidate data. Specifically, the temporary center coordinate is C (x, y), the reference diameter is r, and the angle between the line segment AC and the X axis is θ. In this case, if Expression (5) or Expression (6) holds, the image data of the pixel at the point A is invalidated.
x + r * cos θ <x coordinate of left line (5)
x + r * cos θ> x coordinate of right line (6)

  In this way, by invalidating the image data in the area exceeding the width of the transport path, the center position of the coin can be accurately determined even when foreign objects such as transport pins are imaged outside the transport path. Can be detected.

  As described above, the present embodiment is a center position determination method applied to a center position determination device that detects the center position of a circular object. The center position determination method acquires an image of a circular object and uses the acquired image as a feature of the acquired image. Based on the set temporary center of the circular object, obtain an image obtained by polar-coordinate conversion of the image acquired for the set temporary center, detect the trajectory of the outer periphery of the circular object based on the acquired image, The temporary center of the circular object is corrected to the true center position on the basis of the detected outer track.

  Therefore, in the center position determination method according to the present invention, the center position of the circular object can be detected with high accuracy even when the image of the captured circular object is imaged of dirt or lack of the outer peripheral portion. it can.

  The center position determination device described in the claims of the present invention is the coin processing device 10, the image acquisition step is the image acquisition unit 14a, the temporary center setting step is the temporary center setting unit 14b, and the polar coordinate image acquisition step is the outer periphery cutting unit 14c. The outer periphery detection step is an example of the outer periphery detection unit 14d, and the center correction step is an example of the center correction unit 14f.

  In the above-described embodiment, the image data portion divided into 32 equal parts is extracted from the polar coordinate-converted image data and the outer peripheral portion and the center position are corrected. However, a predetermined number, for example, 64 equal image data portions are extracted. It is good to do.

  Moreover, although the example applied to a coin processing apparatus was demonstrated in the Example mentioned above, the apparatus which test | inspects the formation precision of circular parts, such as a thread | yarn, a washer, etc., or eccentricity. The present invention can be similarly applied to a device for inspecting the degree.

  As described above, the center position determination method and the center position determination device according to the present invention provide a circular object with high accuracy even when the image of the captured circular object includes an image of dirt or a lack of the outer peripheral portion. This is useful when detecting the center position of the.

DESCRIPTION OF SYMBOLS 10 Coin processing apparatus 11 Imaging part 13 Memory | storage part 13a Image information 13b Center position information 13c Outer periphery information 14 Control part 14a Image acquisition part 14b Temporary center setting part 14c Outer periphery part cutout part 14d Outer periphery detection part 14e Outer periphery correction part 14f Center correction part 14g Output section

Claims (10)

A center position determination method applied to a center position determination device for detecting a center position of a circular object,
An image acquisition step of acquiring an image of the circular object;
A temporary center setting step of setting a temporary center of the circular object based on the characteristics of the image acquired by the image acquisition step;
The image acquired by the image acquisition step with respect to the temporary center set by the temporary center setting step so as to include the outer circumference of the circular object having the largest and smallest diameters among a plurality of types of circular objects of which the center position is to be determined. A polar coordinate image acquisition step for acquiring an image obtained by polar-transforming a partial image surrounded by two concentric circles set to
An outer periphery detecting step of detecting a trajectory of the outer peripheral portion of the circular object based on the image acquired by the polar coordinate image acquiring step;
A center position determination method comprising: a center correction step of correcting the temporary center of the circular object to a true center position based on the track of the outer periphery detected by the outer periphery detection step .   The center position determination method according to claim 1, wherein in the temporary center setting step, horizontal and vertical luminance histograms of the image of the circular object are used. Before Kigaishu detection step,
The center position determination method according to claim 1, wherein the outer peripheral portion is detected based on the partial image that has been subjected to polar coordinate conversion by the polar coordinate image acquisition step. The image acquisition step, the central position determination method according to any one of claims 1 to 3, wherein the handle a predetermined region of the obtained circular was outside yelling reflected image as invalid data . The outer periphery detection step includes
An outer periphery correction step of correcting the trajectory by smoothing the detected trajectory,
The center correction step includes
The center position according to any one of claims 1 to 4 , wherein the temporary center of the circular object is corrected to the true center position based on the trajectory corrected by the outer periphery correction step. Decision method.   The center correction step includes
  The distance between the two points at which the phase difference on the orbit of the outer peripheral part is 180 degrees is calculated for a plurality of locations on the outer peripheral part, and the temporary diameter of the circular object is calculated based on the calculated distance The center position determination method according to any one of claims 1 to 5, wherein:   The center position determining method according to claim 6, wherein in the center correction step, a reference diameter length is obtained from an average of a predetermined number satisfying a predetermined condition among the plurality of temporary diameter lengths.   8. The two outer radii are corrected in the outer peripheral correction step by comparing two radii calculated for the outer peripheral portions at both ends of the diameter with half of the reference radial length. The center position determination method described in 1.   The value of the point on the orbit is substituted for a periodic function representing the distance from a predetermined point on the circumference whose center is the true center position and the radius is half the reference length to the temporary center. The shift amount between the true center position and the temporary center is calculated, and the temporary center is corrected to the true center position using the calculated shift amount. Or the center position determination method of 8. A center position determining device for detecting a center position of a circular object,
Image acquisition means for acquiring an image of the circular object;
Temporary center setting means for setting a temporary center of the circular object based on the characteristics of the image acquired by the image acquisition means;
The image acquired by the image acquisition unit with respect to the temporary center set by the temporary center setting unit so as to include an outer periphery of a circular object having a maximum diameter and a minimum diameter among a plurality of types of circular objects whose center position is to be determined. Polar coordinate image acquisition means for acquiring an image obtained by performing polar coordinate conversion on a partial image surrounded by two concentric circles set to
An outer periphery detecting means for detecting a trajectory of the outer peripheral portion of the circular object based on the image acquired by the polar coordinate image acquiring means;
A center position determination device comprising: center correction means for correcting the temporary center of the circular object to a true center position based on the track of the outer periphery detected by the outer periphery detection means .

Images