JP5853848B2 - Image processing apparatus, image processing program, and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing program, and an image processing method.

  Techniques for measuring colors are known for the purpose of skin diagnosis and design. When performing such color measurement, a dedicated colorimeter such as a photoelectric colorimeter or a spectrophotometer can be used. However, since the dedicated colorimeter is an expensive device, it is easy for the end user. It is difficult to use.

  For this reason, a camera or a digital camera mounted on a portable terminal may be used instead of a dedicated colorimeter. In this way, when taking a picture using a camera, the measurement part of the object to be measured for color is affected by the light source such as sunlight or illumination. There is a light / dark difference between the color of the measurement site.

  For this reason, a through-hole for exposing the skin that is the measurement site in the center is used as the skin measurement area, and the image is photographed with a camera using a color patch that is arranged so that multiple color marking areas surround it. A method of correcting the color of the measurement site shown in the captured image is conceivable. Such a technique utilizes the fact that the color patch affected by the light source is reflected in the image in the same manner as the measurement part by performing imaging while the color patch is attached in the vicinity of the measurement part. That is, in the above method, based on the amount of change between the color patch color stored in advance and the color patch color that appears in the image, the color of the measurement site that appears in the image is changed to the color that the measurement site originally has. A correction amount for correction is calculated.

  Before obtaining the amount of change in the above correction, it is reasonable to detect the color patch through-holes in the image, that is, the measurement area first, in order to extract the order of the color patch marking areas in the image. Is.

  At this time, if there are few edges in the background area other than the color patches included in the image, the range including the part where the edges are gathered is regarded as the range where the color patches are reflected, and the area at the center of the range May be detected as a measurement region. However, the background area other than the color patch does not always have few edges, and a pattern such as a logo or decoration may be designed around the color patch. In such a case, since the edges are concentrated on the place where the pattern is designed, it is difficult to detect the measurement region from the degree of gathering of the edges.

JP 2011-186916 A Japanese Patent Application Laid-Open No. 07-139918

  By the way, it is conceivable to use a Hough transform useful for extracting a region having a specific shape from an image for detecting the measurement region.

  However, when the Hough transform is used for the detection of the measurement area, there is a problem that the amount of calculation processing increases. For example, when an elliptical measurement region is detected from an image, conversion is performed using four parameters of the two-dimensional coordinates on the image and the long and short sides of the ellipse. Thus, as the number of parameters handled in the conversion increases, the calculation processing amount also increases. In particular, in the case of a mobile terminal, the limit of hardware performance of the mobile terminal is lower than that of a stationary machine, and as a result, the processing load on the mobile terminal is excessive, which may hinder the operation of existing functions.

  The disclosed technique has been made in view of the above, and an object thereof is to provide an image processing apparatus, an image processing program, and an image processing method capable of reducing the amount of calculation processing executed when a measurement region is detected.

  An image processing apparatus disclosed in the present application includes an acquisition unit that acquires an image. Furthermore, the image processing apparatus includes a grouping unit that groups and separates pixels having similar pixel values between adjacent pixels among the pixels included in the image. Furthermore, the image processing apparatus includes a calculation unit that calculates a connection angle at which each of the contour pixels forming the contour of the grouped region is connected to another contour pixel on the contour of the region. Furthermore, the image processing apparatus includes an extraction unit that extracts a partial contour line from the contour line by dividing the contour line forming the region in a section in which a change in the connection angle of the contour pixels is within a predetermined threshold. Have. Furthermore, the image processing apparatus includes a detection unit that detects a region having the largest number of intersections between the perpendicular lines of the partial contour lines from the grouped regions.

  According to one aspect of the image processing apparatus disclosed in the present application, there is an effect that it is possible to reduce the amount of calculation processing executed when the measurement region is detected.

FIG. 1 is a top view of the color patch according to the first embodiment. FIG. 2 is a block diagram illustrating a functional configuration of the image processing apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of an image in which a color patch is captured. FIG. 4 is a diagram illustrating an example of the grouping result. FIG. 5 is a diagram illustrating an example of the X-direction filter. FIG. 6 is a diagram illustrating an example of the Y-direction filter. FIG. 7 is a diagram showing the angle of the luminance gradient. FIG. 8 is a diagram illustrating a configuration example of connection angle data. FIG. 9 is a diagram illustrating an example of extracting a partial outline. FIG. 10 is a diagram illustrating an example of an image of a partial outline extraction result. FIG. 11 is a diagram illustrating a configuration example of the partial outline data. FIG. 12 is a diagram illustrating an example of an intersection of perpendicular lines of partial contour lines. FIG. 13 is a diagram illustrating an image example of the voting result of the intersection. FIG. 14 is a diagram illustrating an example of the voting result of the intersection. FIG. 15 is a diagram illustrating an example of mapping. FIG. 16 is a diagram illustrating an example of the mapping output. FIG. 17 is a flowchart of an overall process procedure of the image processing apparatus according to the first embodiment. FIG. 18 is a flowchart illustrating a procedure of a connection angle calculation process according to the first embodiment. FIG. 19 is a flowchart illustrating the procedure of the partial outline extraction process according to the first embodiment. FIG. 20 is a flowchart (1) illustrating the procedure of the measurement region detection process according to the first embodiment. FIG. 21 is a flowchart (2) illustrating the procedure of the measurement region detection process according to the first embodiment. FIG. 22 is a diagram for explaining an application example. FIG. 23 is a schematic diagram illustrating an example of a computer that executes an image processing program according to the first and second embodiments.

  Embodiments of an image processing apparatus, an image processing program, and an image processing method disclosed in the present application will be described below in detail with reference to the drawings. Note that this embodiment does not limit the disclosed technology. Each embodiment can be appropriately combined within a range in which processing contents are not contradictory.

[Structure of color patch 1]
First, the structure of the color patch according to the present embodiment will be described. FIG. 1 is a top view of the color patch 1 according to the first embodiment. In the example of FIG. 1, a top view when the display surface of the color patch 1 is viewed from the top with the display surface on which the color marking area is formed as the top and the back surface of the display surface in contact with the object whose color is to be measured as the bottom Indicates. In the following, it is assumed that the color of a person's skin is measured using the color patch 1 in order to provide a skin diagnosis service.

  As shown in FIG. 1, the color patch 1 is attached to or placed on a subject whose color is to be measured, and then is imaged together with the measurement region. This is for reflecting the affected color patch 1.

  Here, in the color patch 1 according to the present embodiment, a through hole 3 penetrating the display surface and the back surface of the color patch 1 is formed. Thus, the through-hole 3 is formed in the central portion of the color patch 1 when the color patch 1 is affixed to the subject whose color is to be measured. This is for exposing as a measurement region. And if a subject's skin is exposed from the through-hole 3, the marking area | regions C1-C16 formed in the display surface of the color patch 1 will be arrange | positioned so that a subject's skin may be surrounded.

  For this reason, in the color patch 1 according to the present embodiment, it is possible to bring the respective marking areas arranged in an annular shape around the through hole 3 to be close to the measurement site without omission. Therefore, the color patch 1 according to the present embodiment can make the portion closer to the measurement site larger than the conventional color patch in which the color markings are two-dimensionally arranged in a matrix.

  Furthermore, in the color patch 1 according to the present embodiment, the marking areas C1, C6, C9, and C14 of the same color are formed on both sides of the display surface with the through hole 3 interposed therebetween. Furthermore, the same marking areas C2, C5, C10, and C13 are also formed on both sides of the display surface with the through hole 3 interposed therebetween. Further, the same marking areas C3, C8, C11, and C16 are also formed on both sides of the display surface with the through hole 3 interposed therebetween. Furthermore, the same marking areas C4, C7, C12 and C15 are also formed on both sides of the display surface with the through hole 3 interposed therebetween.

  The marking areas C1, C6, C9 and C14, the marking areas C2, C5, C10 and C13, the marking areas C3, C8, C11 and C16, and the marking areas C4, C7, C12 and C15 shown in FIG. Is an area on the display surface on which the same color is marked. Among these, the marking area C1 and the marking area C9 are arranged in pairs facing each other. Furthermore, the marking area C6 and the marking area C14 are also arranged in pairs facing each other. In addition, the marking areas C2, C5, C10 and C13, the marking areas C3, C8, C11 and C16 and the marking areas C4, C7, C12 and C15 are also different from the marking areas C1, C6, C9 and C14. Are different, but the mutual positional relationship is the same.

  As described above, the marking areas of the same color are arranged so as to face each other because the color of the marking area of the color patch 1 originally changes and the color of the marking area of the color patch 1 that appears in the image together with the measurement site. This is because an accurate change amount is calculated when the image processing apparatus 10 described later corrects the skin color of the subject appearing in the image using the amount.

  That is, it may be insufficient to reproduce the color of the subject's skin simply by observing the amount of change in one marking area. This is because the influence on the subject's skin by the light source is not necessarily uniform over the entire measurement region. Therefore, in the color patch 1 according to the present embodiment, in order to reflect a non-uniform change in the amount of change, the same color is used at the position of the pair close to both ends of the contour representing the measurement region or the position of the pair equivalent thereto. The marking areas were arranged in pairs. As a result, the paired marking areas among the marking areas of the same color are affected by the light source at positions close to or opposite to both ends of the contour that represents the measurement area. Therefore, it is possible to accurately calculate the amount of change in the color of the portion sandwiched between the paired marking areas among the marking areas of the same color. Furthermore, in the color patch 1 according to the present embodiment, four types of colors and two pairs of marking areas for the same color can be reflected in the image. Therefore, it is possible to cause the image processing apparatus 10 described later to calculate the amount of change reflecting the influence of the light source over the entire measurement region.

  Furthermore, when arranging the marking area | region of the same color as a pair, it is preferable to arrange | position symmetrically about the center of the through-hole 3 as a reference | standard. For example, as shown in FIG. 1, a pair of a marking area C1 and a marking area C9, or a pair of a marking area C6 and a marking area C14 are arranged facing each other with the same size and the same shape. To do. Further, the marking areas C2, C5, C10 and C13, the marking areas C3, C8, C11 and C16, and the marking areas C4, C7, C12 and C15, as well as C1, C6, C9 and C14, are two sets. Are arranged symmetrically. As a result, the amount of change in the marking area adjacent to the measurement area is calculated on the diagonal line of the measurement area, and the amount of change that equally reflects the influence of the label area of the same size and shape on the light source is calculated. It becomes possible.

  As described above, in the color patch 1 according to the present embodiment, it is possible to bring the respective marking areas arranged in an annular shape around the through hole 3 to be close to the measurement site without omission. Further, when the color patch 1 according to the present embodiment is photographed together with the measurement site, the amount of change in the color of the portion sandwiched between the paired labeling areas among the labeling areas of the same color from the photographed image. Can be calculated accurately. Therefore, when correcting the color of the measurement region shown in the image using the amount of change in the color of the marking area arranged in the color patch 1, the probability of reproducing the original color of the measurement region or a color close thereto is increased. be able to. Therefore, according to the color patch 1 according to the present embodiment, the color measurement accuracy can be improved.

  Further, in the color patch 1 according to the present embodiment, white marking areas W1 to W4 are arranged at a predetermined interval, for example, an interval of 90 degrees. For this reason, it is possible to correct the white balance by using the pixel values when the four white marking areas W1 to W4 adjacent to the measurement area appear in the image.

  Furthermore, in the color patch 1 according to the present embodiment, the outer shape of the color patch 1 is formed in an annular shape by forming the shape of the through hole 3 forming the outer periphery of the display surface and the inner periphery of the display surface in a circular shape. As a result, the measurement area photographed together with the measurement site by the camera 11 to be described later can be circular, and the outer shape of the color patch 1 can be circular. For this reason, even if the display surface of the color patch 1 is not photographed from the front but photographed from various directions other than the front, the photographed image displays the measurement area and the color patch 1 on the subject's skin. Surface deformation can be suppressed to an elliptical shape. Therefore, it is possible to easily detect the measurement region and the display surface from the captured image. Further, in the color patch 1 according to the present embodiment, the outer periphery or inner periphery of the color patch 1 is formed in a circle or an ellipse, so that a guide is displayed on the camera finder screen when photographing the color patch 1 and the measurement site. In addition, since the direction of the guide can be displayed constantly regardless of the direction in which the color patch 1 fits in the lens of the camera, the design and usability can be improved.

  In addition, in the color patch 1 according to the present embodiment, the display surface is divided into two layers of an outer ring and an inner ring, and the marking areas C1, C6, C9, and C14, the marking area C2, C5, C10, and C13, marking areas C3, C8, C11, and C16, and marking areas C4, C7, C12, and C15 are arranged. For this reason, it is possible to efficiently arrange a larger number of marking areas as compared with the case where the marking areas are arranged without being divided into an outer ring and an inner ring.

  In addition, the marking areas C1, C6, C9, and C14, the marking areas C2, C5, C10, and C13, the marking areas C3, C8, C11, and C16, and the marking areas C4, C7, C12, and C15 illustrated in FIG. Any color can be adopted as the color to be displayed. Preferably, a color similar to skin color is employed. For example, when a four-color marking area is arranged, a color distributed in a position surrounding an outer edge of an area in which human skin color is generally distributed in RGB (Red-Green-Blue color model) space, for example, skin color A reddish or yellowish color can be used. In addition, the color of the marking area displayed on the display surface of the color patch 1 may be colored by printing, or may be colored by other methods.

  Further, in the example of FIG. 1, the case where the color patch 1 has only a marking area related to color measurement is illustrated, but a logo, a decoration, or a code is displayed outside the display surface of the color patch 1 or the outer periphery of the color patch 1. Etc. may be designed as a design.

  The shape of the through hole 3 of the color patch 1 may be other shapes such as a polygon or a rectangle instead of a circle or an ellipse.

[Configuration of image processing apparatus]
Subsequently, a functional configuration of the image processing apparatus according to the present embodiment will be described. FIG. 2 is a block diagram illustrating a functional configuration of the image processing apparatus according to the first embodiment. An image processing apparatus 10 shown in FIG. 2 performs image processing for detecting a measurement region formed by the through holes 3 of the color patch 1 from images taken by the camera 11 or the like.

  As an aspect of the image processing apparatus 10, the image processing apparatus 10 can be implemented by installing an image processing program for providing the image processing service on a personal computer connected with a digital camera or a mobile terminal such as a smartphone equipped with a camera function. As another aspect, it may be implemented as a Web server that executes the above-described image processing program, or may be implemented as a cloud that provides image processing services through outsourcing.

  As illustrated in FIG. 2, the image processing apparatus 10 includes a camera 11, a storage unit 13, and a control unit 15. Note that the image processing apparatus 10 includes various functional units included in known computers other than the functional units illustrated in FIG. 2, such as various input devices, audio output devices, and communication interfaces for communicating with external devices. It does not matter as having the functional part.

  The camera 11 is an imaging device using a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), or the like. As an implementation example of the camera 11, a digital camera may be connected, or the camera can be used when the camera is mounted in advance, such as a mobile terminal. Although the case where the image processing apparatus 10 includes the camera 11 is illustrated here, the image does not necessarily need to include the camera 11 when an image can be acquired via a network or a storage device.

  The storage unit 13 is a storage device that stores various programs such as an OS (Operating System) executed by the control unit 15 and an image processing program. As one aspect of the storage unit 13, a storage device such as a semiconductor memory element such as a flash memory, a hard disk, or an optical disk can be cited. The storage unit 13 is not limited to the above-mentioned types of storage devices, and may be a RAM (Random Access Memory) or a ROM (Read Only Memory). The storage unit 13 stores connection angle data 13a, partial outline data 13b, voting data 13c, and the like, which will be described later, as an example of data used in a program executed by the control unit 15.

  The control unit 15 has an internal memory for storing programs defining various processing procedures and control data, and executes various processes using these. As shown in FIG. 2, the control unit 15 includes an acquisition unit 15a, a grouping unit 15b, a calculation unit 15c, an extraction unit 15d, a detection unit 15e, an association unit 15f, and an output unit 15g.

  The acquisition unit 15a is a processing unit that acquires an image. As one aspect, the acquisition unit 15a acquires an image photographed by the camera 11. As another uniformity, the acquisition unit 15a can also acquire an image via a network such as the Internet or a LAN (Local Area Network). As a further aspect, the acquisition unit 15a can acquire an image from an auxiliary storage device such as a memory card or a USB (Universal Serial Bus) memory as well as a memory and a hard disk.

  The grouping unit 15b is a processing unit that groups pixels having similar pixel values between adjacent pixels among the pixels included in the image. As one aspect, the grouping unit 15b is configured such that, among pixels included in the image acquired by the acquisition unit 15a, a pixel whose luminance is a predetermined threshold, for example, 100 or more, is an R component and a G component between adjacent pixels. And pixels having similar B components are grouped. As described above, the grouping is performed by narrowing down to pixels that are equal to or greater than the threshold value by excluding the pixels other than the pixels corresponding to the skin that are assumed to be exposed in the indication area and the measurement area of the color patch 1 from the grouping target. This is to improve accuracy and efficiency. Thereafter, the grouping unit 15b assigns an area number for identifying the area to the area obtained as a result of the grouping.

  Here, a specific example of grouping will be described with reference to FIGS. 3 and 4. FIG. 3 is a diagram illustrating an example of an image in which the color patch 1 is captured. FIG. 4 is a diagram illustrating an example of the grouping result. When grouping is performed on the image shown in FIG. 3, 23 areas of area numbers 1 to 23 shown in FIG. 4 are extracted. As shown in FIG. 3, when an image is picked up after the color patch 1 is applied to the skin of the subject person, as shown in FIG. The grouping result can be obtained in a state where irrelevant areas are mixed. In the example of FIG. 4, regions useful for color measurement such as the skin measurement region exposed from the through hole 3 and the marking region of the color patch 1 are extracted as region numbers 2 to 23 (excluding region number 12). The At the same time, an area such as a background that is not related to the color patch 1 is extracted as area number 1 or area number 12.

  The calculation unit 15c is a processing unit that calculates, for each pixel forming the outline of the area, a connection angle at which the pixel is connected to other pixels on the outline of the area. In the following, the line that forms the contour of the region is referred to as “contour line”, and the pixel that forms the contour of the region, in other words, the pixel that forms the contour line is referred to as “contour point”. The number assigned to the contour point may be described as “contour point number”.

  As one aspect, the calculation unit 15c assigns a contour point number to each contour point included in the region grouped by the grouping unit 15b. At this time, the calculation unit 15c can start with an arbitrary contour point and assign contour point numbers in the order in which the contour points are connected to the starting point. For example, in the plane including the image, when the width direction of the image is the X axis and the height direction of the image is the Y axis, the contour point having the minimum value of X or Y can be set as the starting point. A contour point number can be assigned starting from an arbitrary point, for example, a contour point at the shortest distance from the origin. In addition, the calculation unit 15c selects one contour point, and applies the X-direction filter and the Y-direction filter to the selected contour point and peripheral pixels centered on the contour point.

  FIG. 5 is a diagram illustrating an example of the X-direction filter. FIG. 6 is a diagram illustrating an example of the Y-direction filter. As shown in FIG. 5, the X-direction filter is a filter in which 3 × 3 elements are arranged. Zero is applied to the column including the central element, negative value is applied to the left column, and positive is applied to the right column. Is set as a filter coefficient. In addition, as shown in FIG. 6, the Y-direction filter is a filter in which 3 × 3 elements are arranged. Zero is applied to the row including the central element, negative value is applied to the upper row, and lower row is processed. A positive value is set as a filter coefficient. 5 and 6 exemplifies a filter in which 3 × 3 elements are arranged, but an arbitrary number of filters, for example, elements in which 9 × 9, 16 × 16, and the like are arranged may be used. it can.

  When the X-direction filter shown in FIG. 5 is applied, the luminance value of the contour point that is the target pixel is multiplied by the filter coefficient “0” of the center element of the X-direction filter. At the same time, the luminance value of the peripheral pixels around the contour point is multiplied by the filter coefficient of other elements than the central element included in the X-direction filter. For example, the luminance value of each pixel above, upper right, right, lower right, lower, lower left, left, upper left of the contour point is set to the first row, second column, first row, third column, second row, third column, third of the X direction filter. Filter coefficients "0", "1", "2", "1", "0", "-1" for the third row, third row, second row, third row, first column, second row, first column, first row, first column ”,“ −2 ”,“ −1 ”, respectively. Thereafter, the calculation unit 15c calculates the X component of the luminance gradient between the contour point that is the target pixel and the peripheral pixel by calculating the sum of the values obtained by multiplying the contour point and the peripheral pixel by the filter coefficient. To do.

  When the Y-direction filter shown in FIG. 6 is applied, the luminance value of the contour point that is the target pixel is multiplied by the filter coefficient “0” of the center element of the Y-direction filter. At the same time, the luminance values of the peripheral pixels around the contour point are multiplied by the filter coefficients of elements other than the central element included in the Y-direction filter. For example, the luminance value of each pixel above, upper right, right, lower right, lower, lower left, left, upper left of the contour point is set to the first row, second column, first row, third column, second row, third column, third of the Y direction filter. Filter coefficients `` -2 '', `` -1 '', `` 0 '', `` 1 '', `` 2 '', `` 2 '', `` 2 '', `` 2 '', `` 2 '', `` 1 '', `` 2 '', `` 0 '' “1”, “0”, and “−1” are respectively multiplied. Thereafter, the calculation unit 15c calculates the Y component of the luminance gradient between the contour point that is the target pixel and the peripheral pixel by calculating the sum of the values obtained by multiplying the contour point and the peripheral pixel by the filter coefficient. To do.

  After calculating the X and Y components of the brightness gradient, the calculation unit 15c calculates the connection angle of the contour points using the X and Y components of the brightness gradient. FIG. 7 is a diagram showing the angle of the luminance gradient. For example, as illustrated in FIG. 7, the calculation unit 15 c performs a calculation of substituting the values of the X component and the Y component of the luminance gradient into the arctangent function “atan (X component / Y component) * 180 / 3.14”, so-called inverse The angle of the luminance gradient is calculated by taking the tangent. Here, the following description will be made assuming that the angle is represented as 0 degree in the upward direction of the image and clockwise from there.

  Since the direction perpendicular to the angle of the luminance gradient is the connecting direction of the contour line, when the X component of the luminance gradient is a positive value, 90 degrees is added to the angle of the luminance gradient, while the luminance gradient X When the component is a negative value, 270 degrees is added to the angle of the luminance gradient. Thereby, the connecting angle of the contour points is calculated. At this time, when the X component of the luminance gradient is zero, the luminance gradient angle diverges. Therefore, in this case, when the Y component of the luminance gradient is larger than zero without taking the arctangent, the concatenated angle of the contour points is set to 180 degrees, while the Y component of the luminance gradient is less than zero. In this case, the connecting angle of the contour points is set to 0 degree.

  Here, the case where the angle of the luminance gradient is converted into the concatenated angle of the contour point is illustrated, but it is not always necessary to perform conversion, and the difference is used as the change in angle in the subsequent processing, so the angle of the luminance gradient Can also be used as the connecting angle of contour points. In addition, here, an example in which the unit of the connection angle is converted when taking the arc tangent has been described, but it is not always necessary to convert from radians to degrees.

  Thereafter, the calculation unit 15c calculates the connection angle for each contour point included in each region as described above, and stores the connection angle data 13a in which the contour point included in each region is associated with the connection angle. Register to the unit 13. FIG. 8 is a diagram illustrating a configuration example of the connection angle data 13a. In the example of the connection angle data 13a shown in FIG. 8, the contour point numbers of the contour points included in the region having the region number are associated with each region number. Further, in the connection angle data 13a, the coordinates (X, Y) and the connection angle at which the contour point having the contour point number is located on the image are associated with each contour point number. For example, in the case of the region number 1, the coordinates from the contour point number 1 to the contour point number N and the connection angle are associated with the region number 1.

  In the example of FIG. 8 described above, the case where the data in which the region number, the contour point number, the coordinates, and the connection angle are associated is illustrated, but the connection angle data scheme is not limited thereto. For example, the disclosed apparatus can manage the association between the region number and the contour point number, the association between the contour point number and the coordinate, and further the association between the contour point number and the connection angle as separate data.

  Returning to the description of FIG. 2, the extraction unit 15 d is a processing unit that extracts a partial contour line obtained by dividing a contour line that forms a region. As an aspect, the extraction unit 15d divides the contour line at contour points where the change in the connection angle exceeds a predetermined threshold. In other words, the extraction unit 15d separates the contour lines that form the region in a section where the contour points are connected without greatly changing the connection angle.

  Explaining this, the extraction unit 15 d refers to the connection angle data 13 a stored in the storage unit 13. Then, the extracting unit 15d selects one area number included in the connection angle data 13a. Subsequently, the extraction unit 15d selects one of the contour point numbers associated with the region number, and registers the selected contour point number and the connection angle in an internal memory (not shown). As a result, the articulation angle of the contour point that is a candidate for the starting point of the partial contour line is temporarily stored for comparison with the articulation angle of the contour point that is searched in the subsequent processing.

  After that, the extraction unit 15d searches for the contour point number on the contour line of the region, and is temporarily stored as a candidate for the start point of the partial contour line in the internal memory and the connection angle of the contour point having the contour point number. It is determined whether or not the difference between the contour points and the connection angle exceeds a predetermined division threshold. Here, the division threshold value is such that determination can be branched at a connection portion between the concentric circular arc of the through hole 3 and the line segment in the direction toward the through hole 3 in the outline of the marking region of the color patch 1. It is preferable to set a value. In addition, when the shape of the outer periphery or inner periphery of the color patch 1 is an ellipse or a polygon, the division threshold can be appropriately changed according to the shape.

  At this time, when the difference in the connecting angle exceeds the division threshold, the connecting direction of the contour line is large in the section from the contour point that is the candidate for the starting point to the contour point being searched among the contour lines that form the contour of the region. It can be determined that it has changed. In this case, the extraction unit 15d activates the division flag of the contour point searched before the contour point being searched, for example, the immediately preceding contour point, and sets the contour point being searched as the starting point of the partial contour line. The contour point number and the connection angle are overwritten and registered in the internal memory as candidates. For example, the extraction unit 15d initializes a division flag, which is a division point for dividing the contour line on the internal memory, and arranges it for each contour point included in the region, and exceeds the division threshold. Update the contour point division flag to "1". As a result, the partial outline can be extracted as a result of dividing the outline of the region.

  On the other hand, when the difference in the connecting angle is equal to or smaller than the division threshold, the connecting direction of the contour line is large in the section from the contour point that is the candidate for the starting point to the contour point being searched among the contour lines that form the contour of the region. Can be determined as not changing. In this case, the extraction unit 15d searches for the contour point number of the contour point connected to the contour point searched this time among the unsearched contour points until the unsearched contour point disappears. The division determination is repeatedly executed. Then, when there is no unsearched contour point, the extraction unit 15d repeats the above-described division determination until there is no unsearched area, with an area that has not been subjected to the above-described division determination as a search target.

  After that, the extraction unit 15d divides the contour line between the contour point for which the division flag is validated and the contour point searched next after the above-described division determination is completed for all the regions. In other words, the extraction unit 15d includes, from the set of contour point numbers of all the contour points that form the contour line of the region, from the contour point number that is a candidate for the start point of the partial contour line to the contour point number for which the division flag is activated. Are separated as a set of contour point numbers of partial contour lines. After that, the extraction unit 15d registers the set of contour point numbers, coordinates, and connection angles of the contour points forming the partial contour line in the storage unit 13 as the partial contour line data 13b.

  FIG. 9 is a diagram illustrating an example of extracting a partial outline. In the example of FIG. 9, it is assumed that the contour line S that starts the contour point S and searches for the contour line 200 that forms the contour of the region with the contour point E as the end point is assumed. As shown in FIG. 9, when the search is started from the contour point S, a concentric circular arc of the through hole 3 is searched until the search reaches the contour point A. Therefore, the difference in the articulation angle between the contour point S and the contour point being searched remains a slight difference. For this reason, until the search reaches the contour point A, the division threshold is not exceeded in the above-described division determination, and the division flag is not validated. Thereafter, when the search reaches the contour point A, the search for the concentric circular arc of the through hole 3 is completed, and the search for the line segment in the direction toward the through hole 3 is started. Then, the connection angle difference between the contour point S and the contour point A deviates to a right angle. For this reason, when the contour point A is searched, the division threshold is exceeded in the above-described division determination, the contour point division flag searched immediately before the contour point A is validated, and the contour point A becomes a new partial contour. Candidate for the starting point of the line. With this division flag, the contour line 200 is divided into a partial contour line 210 and an unsearched contour line 220.

  Thereafter, until the search reaches the contour point B, a line segment in the direction toward the through hole 3 is searched. Therefore, the difference in the articulation angle between the contour point A and the contour point being searched remains a slight difference. For this reason, until the search reaches the contour point B, the division threshold is not exceeded in the above-described division determination, and the division flag is not validated. Thereafter, when the search reaches the contour point B, the search for the line segment in the direction toward the through hole 3 is completed, and the search for the concentric arc of the through hole 3 is started. Then, the connection angle difference between the contour point A and the contour point B deviates to a right angle. For this reason, when the contour point B is searched, the division threshold is exceeded in the above-mentioned division determination, the contour point division flag searched immediately before the contour point B is validated, and the contour point B becomes a new partial contour. Candidate for the starting point of the line. With this division flag, the contour line 200 is further separated into a partial contour line 230 and an unsearched contour line 240.

  In the section from the contour point B to the contour point E which is the end point, a concentric circular arc of the through hole 3 is searched, so that the difference in the connecting angle between the contour point A and the contour point being searched is slightly different. Stay on. Therefore, the division threshold is not exceeded in the above-described division determination until the end point, and the division flag is not validated. Therefore, the contour line 240 that has not been searched when the contour point B is searched is extracted as a partial contour line as it is. In this way, the partial outline 210, the partial outline 230, and the partial outline 240 are extracted from the outline 200.

  FIG. 10 is a diagram illustrating an example of an image of a partial outline extraction result. For example, when the division determination and the validation of the division flag described with reference to FIG. 9 are executed for the outlines of all regions, the partial outline shown in FIG. 10 is extracted. As shown in FIG. 10, a concentric circular arc of the through hole 3 or a line segment in the direction toward the through hole 3 is extracted as a partial outline from the outline of the marking region of the color patch 1. In addition to this, the contour line of the through hole 3 of the color patch 1, that is, the contour point before the substantially circular contour line on the image exceeds the division threshold, and the contour point after the division threshold is exceeded. As a result, the circular arc of the through hole 3 is extracted as a partial contour line.

  FIG. 11 is a diagram illustrating a configuration example of the partial outline data 13b. In the example of the partial outline data 13b shown in FIG. 11, the partial outline number of the partial outline included in the area having the area number is associated with each area number. Furthermore, in the partial outline data 13b, for each partial outline number, a set of outline point numbers, coordinates, and connection angles of outline points forming a partial outline having the partial outline number is associated. For example, in the case of region number 1, M partial contour lines from partial contour number 1 to partial contour number M are associated with region number 1. Further, the coordinates of the contour points from the contour point number 1 to the contour point number N1 that form the partial contour line of the partial contour number 1 and the connection angle are associated with the region number 1. Further, the coordinates of the contour points from the contour point number 1 to the contour point number N2 that form the partial contour line of the partial contour number 2 and the connection angle are associated with the region number 1. Further, the coordinates of the contour points from the contour point number 1 to the contour point number NM that form the partial contour line of the partial contour number M and the connection angle are associated with the region number 1. The sum of N1 contour points, N2 contour points,..., NM contour points included in M partial contour lines from partial contour number 1 to partial contour number M is a region It is equal to the number of all contour points included in the area of number 1.

  Returning to the description of FIG. 2, the detection unit 15 e is a processing unit that detects an area where the intersections of the perpendicular lines of the partial contour lines are the largest as the measurement area of the color patch 1. In this way, the area having the largest number of intersections between the perpendicular lines of the partial contour lines is regarded as the measurement area. When voting is performed on the area where the intersection points between the perpendicular lines of the partial contour lines exist, the number of votes in the measurement area is the largest. This is because the probability of becoming is high.

  This is because, when voting is performed for an area where there is an intersection of perpendicular lines between partial contour lines such as an arc of the through hole 3 of the color patch 1 or an arc of a concentric circle of the through hole 3, the number of votes is the measurement of the color patch 1. It is because it concentrates on the through-hole 3 which is a region. On the other hand, voting was started for other partial contour lines, for example, a line segment in the direction toward the through-hole 3, and an area where a perpendicular intersection exists between the partial contour lines forming the logo or decoration provided on the color patch 1. In some cases, the number of votes is not concentrated in a specific area but dispersed. Therefore, when voting is performed on a region where the intersections of the perpendiculars of the partial contour lines exist between the partial contour lines, there is a high probability that the voting for the measurement region is the most. From this, the detection unit 15 e detects the area where the intersections of the perpendicular lines of the partial contour lines are the largest as the measurement area of the color patch 1.

  FIG. 12 is a diagram illustrating an example of an intersection of perpendicular lines of partial contour lines. As illustrated in FIG. 12, the detection unit 15 e calculates a perpendicular 315 of the partial contour 310 by calculating a straight line perpendicular to a line segment connecting the contour points at both ends among the contour points included in the partial contour 310. . The detection unit 15 e also calculates the perpendicular 325 of the partial contour 320 by calculating a straight line that is perpendicular to the line connecting the contour points at both ends among the contour points included in the partial contour 320. Thereafter, the detection unit 15e discriminates an area where the intersection point P between the perpendicular line 315 of the partial outline 310 and the perpendicular line 325 of the partial outline 320 exists, and then counts up the number of intersections in the area where the intersection point P exists. For example, the detection unit 15e calculates the number of intersections corresponding to the area where the intersection calculated earlier is included in the voting data 13c stored in the storage unit 13, that is, the data in which the number of intersections of the area is associated with each area. Update to add. And the detection part 15e performs the calculation of an intersection, and the count of the number of intersections about all the combinations of each partial outline. In addition, although the case where the perpendicular line of the partial contour line is calculated using the contour points at both ends of the partial contour line is illustrated here, the normal line passing through the midpoint of the curve forming the partial contour line is defined as the perpendicular line of the partial contour line. Can also be calculated as

  FIG. 13 is a diagram illustrating an image example of the voting result of the intersection. For example, when voting is performed on an area where intersections exist for all combinations of partial contour lines, the voting result shown in FIG. 13 can be obtained. At this time, the partial contour line includes not only the arc of the through-hole 3 of the color patch 1 and the concentric arc of the through-hole 3, but also a line segment in the direction toward the through-hole 3 and a logo or decoration provided on the color patch 1. This includes the edges. For this reason, as shown in FIG. 13, some voting is also performed in an area other than the measurement area of the color patch 1, but most of the voting is performed in the measurement area of the color patch 1.

  In addition, the detection unit 15 e detects the region where the intersections of the perpendicular lines of the partial contour lines are the largest as the measurement region of the color patch 1. FIG. 14 is a diagram illustrating an example of the voting result of the intersection. For example, when the voting result shown in FIG. 14 is obtained, the number of intersections voted for the region number 11 out of the region numbers 1 to 23 obtained as the grouping result shown in FIG. 4 is “1489”. , Will be the most. Therefore, the detection unit 15 e detects the area number 11 having the largest number of intersections as the measurement area of the color patch 1.

  Returning to the description of FIG. 2, the association unit 15 f uses the detection result of the measurement area of the color patch 1 to arrange the area number of the marking area surrounding the measurement area and the outline of the measurement area when it circulates clockwise. A processing unit that associates the order.

  As one aspect, for each contour point of each contour point that forms the contour of the measurement region on the image, the associating unit 15f has a corresponding contour that is the closest to the contour point of the measurement region among the contour points of each region. A point is extracted, and the contour point number of the contour point of the measurement region is assigned to the corresponding contour point. Then, each time the corresponding contour point is extracted, the associating unit 15f increments the number of corresponding contour points of the region for the region from which the corresponding contour point is extracted, and the corresponding contour point of the region. Update is performed in which the contour point number assigned to the corresponding contour point extracted this time is added to the total value of the contour point numbers assigned to. Subsequently, when the corresponding contour points are extracted over all the contour points in the measurement region, the associating unit 15f performs a calculation of dividing the total value of the contour point numbers assigned to the corresponding contour points by the number of corresponding contour points. By executing this, an average value of contour point numbers for each region number is calculated as a representative value for each region. After that, the associating unit 15f associates the region numbers in order from the region number of the region with the small representative value. Thereby, the arrangement order of the area numbers of the proximity areas close to the measurement area is obtained.

  Thereafter, the associating unit 15f associates the arrangement order of the outer areas existing outside the proximity area with the area numbers of the areas in accordance with the arrangement order of the area numbers of the proximity areas. Thereby, the arrangement order of the area numbers of the outer area is obtained. Then, the associating unit 15f determines that the arrangement order in which the distance from the center coordinate of the measurement region to the outer edge of the proximity region is the longest is a penetration region, that is, a white marking region. Here, in the example of the color patch 1, there is a regularity that the white marking area that is the penetrating area is arranged so as to skip two arrangement orders. Accordingly, the association unit 15f further determines that the arrangement order L + 3, the arrangement order L + 6, and the arrangement order L + 9 are further through areas in addition to the arrangement order L that has been previously determined as the through area.

  The output unit 15g is a processing unit that outputs the RGB value of each marking region of the color patch 1 to a predetermined output destination using the arrangement order of the region numbers of the proximity region and the outer region and the discrimination result of the penetration region. Here, in the following, it is assumed that the output format of the data to the output destination is a format that outputs the RGB values of the white marking areas W1 to W4 and the RGB values of the marking areas C1 to C16 of colors other than white. Furthermore, it is assumed that the luminance of the marking area C1 is higher than the luminance of the marking area C2. Note that the above RGB values are average values of the RGB values of the pixels constituting each marking area.

  Since the output unit 15g performs mapping between the area numbers associated with the arrangement order, the white marking areas W1 to W4, and the marking areas C1 to C16 of colors other than white, the mapping is performed in the mapping order. Specify the order of output.

  Explaining this, the color patch 1 includes the inner and outer marking areas sandwiched between the penetrating area of the color patch 1 and the penetrating area adjacent to the 90-degree area with the color of the inner marking area every 90 degrees. If it is arranged so that the colors of the outer marking areas are switched, there is regularity. By utilizing this, the output unit 15g has one adjacent order region that is larger in the order of arrangement than the arrangement order L determined as the through region in the order of arrangement of the adjacent region and the outer region, that is, one right side of the through region. It is determined whether or not the luminance of a certain adjacent region is larger than the luminance of the outer region. The reason why the adjacent area and the outer area adjacent to each other are compared in this manner is that the influence of the light source is small when viewed locally. Further, the luminance is compared because the luminance is less affected by the light source than the RGB value, and it is not necessary to assume a case where the magnitude relationship between the luminance of the adjacent area and the outer area is reversed by shooting.

  When the brightness of the proximity area is higher than the brightness of the outside area, it can be determined that the order of arrangement of the proximity area and the outside area corresponds to the order of arrangement of the marking areas defined in the data output format. In this case, the output unit 15g starts the arrangement order L of the penetrating areas as a starting point without shifting the arrangement order, and assigns the area numbers associated with each arrangement order to the white marking areas W1 to W4 and the white color. It maps with the marking area | regions C1-C16 of colors other than. On the other hand, when the brightness of the proximity area is less than or equal to the brightness of the outside area, it can be determined that the order of arrangement of the proximity area and the outside area deviates from the order of alignment of the marking areas defined in the data output format. In this case, the output unit 15g shifts the arrangement order so that the arrangement order L + 3 of the through area one preceding the arrangement order L of the through areas becomes the arrangement order L. Thereby, the output unit 15g shifts the original arrangement order L + 3 to the arrangement order L, shifts the original arrangement order L + 6 to the arrangement order L + 3, shifts the original arrangement order L + 9 to the arrangement order L + 6, and restores the original arrangement order. The order L is shifted to the order L + 9. In this way, the color patches 1 appearing in the image are shifted in the arrangement order when they are rotated 90 degrees. Thereafter, the output unit 15g starts from the arrangement order L of the through areas after the shift, and uses the area numbers associated with each arrangement order as the white indication areas W1 to W4 and the indication areas C1 to C16 of colors other than white. And mapping.

  For example, an area number in the arrangement order L that is the first element of the penetrating area is mapped to the marking area W1. In addition, the area number of the arrangement order L + 3 which is the second element of the penetrating area is mapped to the marking area W2. In addition, the area number of the arrangement order L + 6 that is the third element of the penetrating area is mapped to the marking area W3. Further, the area number of the arrangement order L + 9, which is the fourth element of the penetrating area, is mapped to the marking area W4. These marking areas W1 to W4 are white marking areas that are penetration areas. For this reason, since there are no outer areas in the marking areas W1 to W4, the area numbers of the adjacent areas of the arrangement order L, the arrangement order L + 3, the arrangement order L + 6, and the arrangement order L + 9 are mapped.

  Furthermore, the area number of the adjacent area of the arrangement order L + 1 that is one order ahead of the arrangement order L that is the first element of the penetrating area is mapped to the marking area C1. Further, the area number of the outer area of the marking area C1 is mapped to the marking area C2. In addition, the area number of the adjacent area in the arrangement order that is one order ahead of the arrangement order of the indication area C1 is mapped to the indication area C3. Further, the area number of the outer area of the marking area C3 is mapped to the marking area C4. Here, since the labeling area W2 is sandwiched between the labeling area C5 and the labeling area C3, the area numbers of the adjacent areas that are two orders ahead of the ordering order of the labeling area C3 are mapped. Further, the area number of the outer area of the marking area C5 is mapped to the marking area C6. In addition, the area number of the adjacent area in the arrangement order that is one order ahead of the arrangement order of the indication area C6 is mapped to the indication area C7. Further, the area number of the outer area of the marking area C7 is mapped to the marking area C8. Here, since the labeling area W3 is sandwiched between the labeling area C9 and the labeling area C7, the area numbers of the adjacent areas that are two orders ahead of the ordering order of the labeling area C7 are mapped. Further, the area number of the outer area of the marking area C9 is mapped to the marking area C10. In addition, the area number of the adjacent area in the arrangement order one order ahead of the arrangement order of the indication area C9 is mapped to the indication area C11. Further, the area number of the outer area of the marking area C11 is mapped to the marking area C12. Here, since the marking area W4 is sandwiched between the marking area C13 and the marking area C11, the area numbers of the adjacent areas in the order of arrangement two orders ahead of the order of the marking area C11 are mapped. Further, the area number of the outer area of the marking area C13 is mapped to the marking area C14. In addition, the area number of the adjacent area in the arrangement order that is one order ahead of the arrangement order of the indication area C13 is mapped to the indication area C15. Further, the area number of the outer area of the marking area C15 is mapped to the marking area C16.

  Here, a specific example of mapping will be described with reference to FIG. FIG. 15 is a diagram illustrating an example of mapping. As shown in FIG. 15, the area of area number 2 shown in FIG. 4 is mapped to the white marking area W <b> 1 of the color map 1. Further, the area of area number 14 shown in FIG. 4 is mapped to the white marking area W2 of the color map 1. Further, the area of area number 21 shown in FIG. 4 is mapped to the white marking area W3 of the color map 1. Further, the area of area number 13 shown in FIG. 4 is mapped to the white marking area W4 of the color map 1. 4 is mapped to the color marking area C1 of the color map 1. Furthermore, the area of area number 4 shown in FIG. 4 is mapped to the color marking area C2 of the color map 1. Further, the area of area number 10 shown in FIG. 4 is mapped to the color marking area C3 of the color map 1. Further, the area number 8 shown in FIG. 4 is mapped to the color marking area C4 of the color map 1. Further, the area of area number 17 shown in FIG. 4 is mapped to the color marking area C5 of the color map 1. Further, the area of area number 18 shown in FIG. 4 is mapped to the color marking area C6 of the color map 1. Further, the area of area number 20 shown in FIG. 4 is mapped to the color marking area C7 of the color map 1. Further, the area of area number 23 shown in FIG. 4 is mapped to the color marking area C8 of the color map 1. Furthermore, the area of area number 19 shown in FIG. 4 is mapped to the color marking area C9 of the color map 1. Further, the area of area number 22 shown in FIG. 4 is mapped to the color marking area C10 of the color map 1. Further, the area of area number 15 shown in FIG. 4 is mapped to the color marking area C11 of the color map 1. Further, the area of area number 16 shown in FIG. 4 is mapped to the color marking area C12 of the color map 1. Furthermore, the area of area number 9 shown in FIG. 4 is mapped to the color marking area C13 of the color map 1. Further, the area of area number 7 shown in FIG. 4 is mapped to the color marking area C14 of the color map 1. Further, the area of area number 5 shown in FIG. 4 is mapped to the color marking area C15 of the color map 1. Further, the area of area number 3 shown in FIG. 4 is mapped to the color marking area C16 of the color map 1.

  Thereafter, the output unit 15g calculates the average value of the RGB values of the pixels included in the marking area for each of the white marking areas W1 to W4 and the marking areas C1 to C16 of colors other than white. In addition, the output unit 15g outputs the RGB values calculated for each of the white marking areas W1 to W4 and the marking areas C1 to C16 of colors other than white to a predetermined output destination. At this time, the output unit 15g can also output the RGB values inherent in the color of each marking area together with the RGB values of each marking area.

  FIG. 16 is a diagram illustrating an example of the mapping output. As shown in FIG. 16, the RGB values of the pixels included in the marking area for each of the mapped white marking areas W1 to W4 and the marking areas C1 to C16 of colors other than white shown in FIG. The average value of is calculated. In addition, together with the average value of the RGB values calculated for each of the white marking areas W1 to W4 and the marking areas C1 to C16 of colors other than white, the RGB values inherent in the colors of each marking area are Is output.

  As an example of the output destination, the RGB value of the measurement region that appears in the image using the RGB value inherent in the color of the indication region of the color patch 1 and the amount of change in the RGB value of the color of the indication region that appears in the image is used. There is a color measurement program to be corrected. Such a color measurement program does not necessarily need to be executed in the image processing apparatus 10, and may be output to an external device in which the color measurement program is executed.

  Note that various integrated circuits and electronic circuits can be employed for the control unit 15. In addition, a part of the functional unit included in the control unit 15 may be another integrated circuit or an electronic circuit. For example, an ASIC (Application Specific Integrated Circuit) is an example of the integrated circuit. Examples of the electronic circuit include a central processing unit (CPU) and a micro processing unit (MPU).

[Process flow]
Next, a processing flow of the image processing apparatus 10 according to the present embodiment will be described. Here, after (1) the entire process executed by the image processing apparatus 10 is described, (2) a connection angle calculation process, (3) a partial contour extraction process, which is executed as a sub-flow of the entire process, (4) The measurement area detection processing will be described in this order.

(1) Overall Processing FIG. 17 is a flowchart illustrating a procedure of overall processing of the image processing apparatus 10 according to the first embodiment. The entire process is started when an image is acquired by the acquisition unit 15a. As shown in FIG. 17, when an image is acquired (step S101), the image processing apparatus 10 groups pixels having similar pixel values between adjacent pixels among the pixels included in the image (step S102). .

  Subsequently, the image processing apparatus 10 performs a “joint angle calculation process” for calculating a joint angle at which the contour point is connected to another contour point on the contour of the region (step S110). S103).

  Then, the image processing apparatus 10 executes “partial contour line extraction processing” for extracting a partial contour line obtained by dividing the contour line forming the region, using the connection angle of each contour point calculated in step S103. (Step S104).

  Subsequently, the image processing apparatus 10 executes “measurement region detection processing” in which the region having the largest number of intersections between the perpendiculars of the partial contour lines extracted in step S104 is detected as the measurement region of the color patch 1 (step S105). ).

  After that, the image processing apparatus 10 uses the detection result of the measurement area of the color patch 1, the area numbers of the adjacent area and the outer area surrounding the measurement area, and the arrangement order when the outline of the measurement area is rotated clockwise. Are associated with each other (step S106).

  The image processing apparatus 10 then arranges the arrangement order L having the longest distance from the center coordinates (CX, CY) of the measurement area to the outer edge of the proximity area, and the arrangement orders L + 3 and L + 6 in the arrangement order and a certain rule. And L + 9 are determined to be through regions (step S107).

  Subsequently, the image processing apparatus 10 determines whether the luminance of the adjacent region having a larger arrangement order than the arrangement order L determined as the through region in the arrangement order of the adjacent region and the outer region is larger than the luminance of the outer region. It is determined whether or not (step S108).

  At this time, when the brightness of the proximity area is equal to or lower than the brightness of the outside area (No in step S108), the arrangement order of the proximity area and the outside area is the alignment order of the marking areas defined in the data output format. It can be determined that there is a gap. In this case, the image processing apparatus 10 shifts the arrangement order so that the arrangement order L + 3 of the penetrating area one prior to the arrangement order L of the penetration areas becomes the arrangement order L (step S109). In addition, the image processing apparatus 10 starts from the arrangement order L of the through areas after the shift, and uses the area numbers associated with each arrangement order as the white indication areas W1 to W4 and the indication areas of colors other than white. Mapping is performed with C1 to C16 (step S110).

  On the other hand, when the brightness of the adjacent area is higher than the brightness of the outer area (Yes in step S108), the arrangement order of the adjacent area and the outer area corresponds to the alignment order of the marking areas defined in the data output format. Then it can be determined. In this case, the image processing apparatus 10 does not shift the arrangement order, but uses the arrangement order L of the penetrating areas as a starting point, and assigns the area numbers associated with each arrangement order to the white marking areas W1 to W4 and Mapping is performed with the marking areas C1 to C16 of colors other than white (step S110).

  Thereafter, the image processing apparatus 10 outputs a predetermined output together with the RGB values of the respective marking areas W1 to W4 and C1 to C16 together with the RGB values of the respective marking areas W1 to W4 and C1 to C16. The data is output first (step S111), and the process is terminated.

(2) Connection Angle Calculation Processing FIG. 18 is a flowchart illustrating the procedure of the connection angle calculation processing according to the first embodiment. This calculation process is a process corresponding to step S103 shown in FIG. 17, and is started after an area is obtained by performing grouping based on closeness of colors.

  As shown in FIG. 18, the image processing apparatus 10 performs initialization for setting an initial value, for example, a number starting with the region number, to the counter A that counts the region numbers (step S <b> 201). Then, the image processing apparatus 10 repeats the processing from step S203 to step S217 described below until the value of the region number counter A exceeds the number of all regions, that is, the last region number (step S202, Yes). Run.

  In other words, the image processing apparatus 10 assigns a contour point number to each contour point included in the region having the value of the region number of the counter A (step S203). Then, the image processing apparatus 10 executes initialization for setting an initial value, for example, a number at which the contour point number starts, to the counter B that counts the contour point number (step S204).

  Thereafter, the image processing apparatus 10 continues to step S206 described below until the value of the contour point number of the counter B exceeds the number of all contour points included in the region A, that is, the last contour point number (Yes in step S205). The processes up to step S216 are repeatedly executed.

  Explaining this, the image processing apparatus 10 applies the X direction filter and the Y direction filter to the contour point having the contour point number of the counter B and the peripheral pixels centered on the contour point (step S206). Thereby, the X component and the Y component of the luminance gradient are obtained. Then, the image processing apparatus 10 determines whether or not the X component of the luminance gradient obtained in step S206 is zero (step S207).

  At this time, when the X component of the luminance gradient is not zero (No in step S207), the image processing apparatus 10 calculates the angle of the luminance gradient by calculating the arc tangent of the values of the X component and the Y component of the luminance gradient. Calculate (step S208).

  When the X component of the luminance gradient takes a positive value (step S209, Yes), the image processing apparatus 10 calculates the concatenated angle of the contour points by adding 90 degrees to the angle of the luminance gradient ( Step S210). On the other hand, when the X component of the luminance gradient takes a negative value (No in step S209), the image processing apparatus 10 calculates the joint angle of the contour points by adding 270 degrees to the luminance gradient angle ( Step S211).

  When the X component of the luminance gradient is zero (step S207, Yes), the angle of the luminance gradient diverges when the arc tangent is taken. Therefore, in this case, the image processing apparatus 10 does not take an arc tangent. That is, when the Y component of the luminance gradient is greater than zero (step S212, Yes), the image processing apparatus 10 determines the connection angle of the contour points as 180 degrees (step S213). On the other hand, when the Y component of the luminance gradient is equal to or less than zero (step S212, No), the image processing apparatus 10 determines the articulation angle of the contour points as 0 degrees (step S214).

  The image processing apparatus 10 stores the concatenated angle of the contour points calculated in step S210, S211, S213, or S214 in association with the area number of the counter A, the contour point number of the counter B, and the coordinates of the contour point. 13 is registered (step S215).

  Thereafter, the image processing apparatus 10 increments the value of the contour point number of the counter B by one (step S216). When the contour point number value of the counter B exceeds the number of all contour points included in the region A (No in step S205), the image processing apparatus 10 increments the value of the region number of the counter A by one ( Step S217). Thereafter, when the value of the area number of the counter A exceeds the number of all areas (No in step S202), the process is terminated.

(3) Partial Contour Line Extraction Processing FIG. 19 is a flowchart illustrating a procedure of partial contour line extraction processing according to the first embodiment. This extraction process is a process corresponding to step S104 shown in FIG. 17, and the process is started after the connection angles of the contour points included in each region are calculated.

  As shown in FIG. 19, the image processing apparatus 10 performs initialization for setting an initial value, for example, a number at which the region number starts, to the counter A that counts the region number (step S <b> 301). Then, the image processing apparatus 10 repeats the processing from step S303 to step S310 described below until the value of the area number counter A exceeds the number of all areas, that is, the last area number (step S302, Yes). Run.

  In other words, the image processing apparatus 10 includes a division flag arranged corresponding to each contour point included in the region having the region number held in the counter A, and a joint angle between contour points that are candidates for the starting point of the partial contour line. Is initialized (step S303). For example, the division flag of each contour point included in the region of region number A is set to an initial value, for example, “0”, and the first contour point number is set to the concatenated angle of the contour points that are candidates for the starting point of the partial contour line. Initialization is performed to set the articulation angle.

  Subsequently, the image processing apparatus 10 executes initialization for setting an initial value, for example, the next contour point number following the first contour point number, to the counter B that counts the contour point numbers (step S304). Thereafter, the image processing apparatus 10 continues to step S306 described below until the value of the contour point number of the counter B exceeds the number of all contour points included in the region A, that is, the last contour point number (Yes in step S305). The processes up to step S309 are repeatedly executed.

  Explaining this, the image processing apparatus 10 has a predetermined difference between the connection angle of the contour point having the contour point number of the counter B and the connection angle of the contour point held as a candidate for the start point of the partial contour line in the register. It is determined whether or not the division threshold is exceeded (step S306).

  At this time, when the difference in the connection angle exceeds the division threshold (step S306, Yes), the contour is contoured in a section from the contour point that is a candidate for the starting point to the contour point being searched among the contour lines forming the contour of the region. It can be determined that the connecting direction of the lines has changed greatly.

  In this case, the image processing apparatus 10 validates the division flag of the contour point with the contour point number B-1 immediately before the contour point number B among the division flags held in the register (step S307). Further, the image processing apparatus 10 overwrites the connection angle of the contour points, which are candidates for the start points of the partial contour lines, held in the register with the connection angle of the contour points having the contour point number of the counter B (step S308).

  On the other hand, when the difference in the connection angle is equal to or smaller than the division threshold (No in step S306), the contour is defined in a section from the contour point that is a candidate for the starting point to the contour point being searched out of the contour lines forming the contour of the region. It can be determined that the connecting direction of the lines has not changed significantly. In this case, the process proceeds to step S309 without updating the division flag and the connection angle in the register.

  Thereafter, the image processing apparatus 10 increments the value of the contour point number of the counter B by one (step S309). When the value of the contour point number of the counter B exceeds the number of all contour points included in the region A (No in step S305), the image processing apparatus 10 increments the value of the region number of the counter A by one ( Step S310).

  When the value of the area number of the counter A exceeds the number of all areas (step S302, No), the image processing apparatus 10 executes the following process. In other words, the image processing apparatus 10 performs the process from the contour point number that is a candidate for the start point of the partial contour line to the contour point number for which the division flag is activated in the set of contour point numbers of all contour points that form the contour line of the region. Are extracted as a set of contour point numbers of partial contour lines (step S311).

  After that, the image processing apparatus 10 registers the set of contour point numbers, coordinates, and connection angles of the contour points forming the partial contour line in the storage unit 13 as the partial contour line data 13b (step S312), and ends the processing. To do.

(4) Measurement Area Detection Process FIGS. 20 and 21 are flowcharts illustrating the procedure of the measurement area detection process according to the first embodiment. This detection process is a process corresponding to step S105 shown in FIG. 17, and is started after the partial outline is extracted.

  As shown in FIG. 20, the image processing apparatus 10 executes initialization to set an initial value, for example, zero which is an unvoted value, in a register that holds the number of intersections for each region (step S401). Subsequently, the image processing apparatus 10 sets an initial value, for example, a number at which the region number starts, to the counter A1 that counts the region number of the first region to which the first partial contour line belongs, out of the two partial contour lines that define the intersection. The initialization for setting is executed (step S402).

  The image processing apparatus 10 then performs step S404 described below until the value of the area number of the first area held in the counter A1 exceeds the number of all areas, that is, the last area number (Yes in step S403). The processes up to step S419 are repeatedly executed.

  Among these, in step S404, the image processing apparatus 10 executes initialization for setting an initial value, for example, the first partial contour line number, to the counter B1 that counts the partial contour line number of the first partial contour line.

  Thereafter, the image processing apparatus 10 until the value of the partial outline number of the first partial outline held in the counter B1 exceeds the number of all partial outlines included in the first area held in the counter A1. (Step S405, Yes), Steps S406 to S418 described below are repeatedly executed.

  Among these, in step S406, an initial value, for example, a number at which the region number starts, is set in the counter A2 that counts the region number of the second region to which the second partial contour line belongs among the two partial contour lines that define the intersection. Perform initialization to

  Thereafter, the image processing apparatus 10 performs processing from step S408 to step S417 described below until the value of the area number of the second area held in the counter A2 exceeds the number of all areas (step S407, Yes). Repeatedly.

  Among these, in step S408, the image processing apparatus 10 executes initialization for setting an initial value, for example, the first partial contour line number, to the counter B2 that counts the partial contour line number of the second partial contour line.

  Thereafter, the image processing apparatus 10 until the value of the partial contour number of the second partial contour line held in the counter B2 exceeds the number of all partial contour lines included in the second region held in the counter A2. (Step S409, Yes), the processes from Step S410 to Step S415 described below are repeatedly executed.

  Explaining this, as shown in FIG. 21, the image processing apparatus 10 has the same area number held in the counter A1 and the counter A2, and the same partial outline number held in the counter B1 and the counter B2. It is determined whether or not (step S410). That is, it is determined whether or not the partial contour lines for which the intersection is to be obtained are the same partial contour lines.

  At this time, if the partial contour lines for which the intersection is to be obtained are not the same partial contour line (step S410, No), the image processing apparatus 10 executes the following processing. That is, the image processing apparatus 10 calculates the coordinates where the intersection of the perpendicular of the first partial outline and the perpendicular of the second partial outline exists (step S411). In addition, when the partial outline which is going to obtain | require an intersection is the same partial outline (step S410, Yes), the process from step S411 to S414 is skipped and it transfers to step S415.

  The “first partial contour line” here refers to a partial contour line included in the first region having the region number of the counter A1 and having the partial contour number of the counter B1. The “second partial contour line” refers to a partial contour line included in the second region having the region number of the counter A2 and having the partial contour number of the counter B2.

  Then, the image processing apparatus 10 determines a region where the intersection calculated in step S411 exists (step S412). Subsequently, the image processing apparatus 10 determines whether or not the intersection is on a region that does not include the edge of the image, for example, a region that does not include the pixels on the four sides that form the image (step S413).

  At this time, if the intersection is on an area not including the edge of the image (step S413, Yes), the image processing apparatus 10 has determined that the intersection exists in step S412 out of the number of intersections held in the register. The number of intersections of the areas is incremented (step S414). If the intersection is on the area including the edge of the image (step S413, No), the process proceeds to step S415 without incrementing the number of intersections.

  Then, the image processing apparatus 10 increments the partial contour line number of the second partial contour line held in the counter B2 by 1 (step S415), and proceeds to the above step S409.

  Thereafter, when the value of the partial outline number of the second partial outline held in the counter B2 exceeds the number of all partial outlines included in the area A2 (No in step S409), the image processing apparatus 10 The process like this is executed. That is, the image processing apparatus 10 increments the value of the area number of the second area held in the counter A2 by 1 (step S417), and proceeds to step S407.

  Then, when the value of the area number of the second area held in the counter A2 exceeds the number of all areas (step S407, No), the image processing apparatus 10 executes the following process. That is, the image processing apparatus 10 increments the partial outline number of the first partial outline held in the counter B1 by 1 (step S418), and proceeds to the above step S405.

  Thereafter, when the value of the partial contour number of the first partial contour line held in the counter B1 exceeds the number of all partial contour lines included in the region A1 (step S405, No), the image processing apparatus 10 The process like this is executed. That is, the image processing apparatus 10 increments the value of the area number of the first area held in the counter A1 by 1 (step S419), and proceeds to step S403.

  When the value of the area number of the first area held in the counter A1 exceeds the number of all areas (step S403, No), the image processing apparatus 10 executes the following process. That is, the image processing apparatus 10 detects the area having the largest number of intersections among the number of intersections held for each area in the register as the measurement area of the color patch 1 (step S420), and ends the process.

[Effect of Example 1]
As described above, the image processing apparatus 10 according to the present embodiment divides the contour line of each region obtained by grouping the pixels in the image in which the color patch 1 is captured with a color within a range in which the change in the connection angle is small, and the partial contour. An area having the largest number of votes at the intersection of the perpendicular lines is detected as a color measurement area. For this reason, the image processing apparatus 10 according to the present embodiment can detect the measurement region of the color patch 1 without performing a circle or an ellipse as in the case of using the Hough transform.

  Therefore, according to the image processing apparatus 10 according to the present embodiment, it is possible to reduce the amount of calculation processing executed when the measurement area is detected. Furthermore, since the image processing apparatus 10 according to the present embodiment can detect the measurement area of the color patch 1 regardless of the presence or absence of a boundary line that divides the marking area, versatility can also be improved.

  Although the embodiments related to the disclosed apparatus have been described above, the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

[Application example]
The disclosed apparatus can also determine whether the positional relationship between the intersection and the partial contour is appropriate after calculating the intersection between the perpendicular of the partial contour and the perpendicular of the partial contour. FIG. 22 is a diagram for explaining an application example. As shown in FIG. 22, the disclosed apparatus calculates a perpendicular 415 of the partial contour 410 by calculating a straight line perpendicular to a line segment 413 connecting the contour points at both ends of the partial contour 410. The disclosed apparatus also calculates the perpendicular line 425 of the partial contour line 420 by calculating a straight line perpendicular to the line segment 423 connecting the contour points at both ends of the partial contour line 420. Then, the disclosed apparatus calculates the coordinates of the intersection point P between the perpendicular line 415 and the perpendicular line 425.

  In addition, the disclosed apparatus performs the following determination on the partial outline 410 and the partial outline 240 that form the intersection P. The disclosed apparatus calculates the number of intersection points in the region where the intersection point exists when both the positional relationship between the intersection point P and the partial contour line 410 and the positional relationship between the intersection point P and the partial contour line 420 are valid. Update.

  Specifically, in the disclosed apparatus, the distance from the intersection point P to the midpoint of the partial contour line 410 is the distance from the intersection point P to the middle point of the line segment 413 connecting the contour pixels at both ends of the partial contour line 410. It is judged whether it is larger than. In the example of FIG. 22, the distance from the intersection point P to the middle point of the partial contour line 410 is larger than the distance from the intersection point P to the middle point of the line segment 413 connecting the contour pixels at both ends of the partial contour line 410. It can be estimated that the positional relationship with the partial contour line is appropriate. Then, the disclosed apparatus determines whether the distance from the intersection point P to the midpoint of the partial contour line 420 is greater than the distance from the intersection point P to the middle point of the line segment 423 connecting the contour pixels at both ends of the partial contour line 420. Determine. In the example of FIG. 22, the distance from the intersection point P to the midpoint of the partial contour line 420 is equal to or less than the distance from the intersection point P to the middle point of the line segment 423 connecting the contour pixels at both ends of the partial contour line 420. It can be estimated that the positional relationship with the partial contour is not valid. As described above, when the positional relationship between the partial contour line and the intersection point is not valid, the number of intersection points in the region where the intersection point exists is not updated. Note that the processing for determining the validity of the above positional relationship can be easily implemented by executing it after step S411 in the flowchart shown in FIG.

  This makes it possible to more accurately detect the measurement area of the color patch 1 even when there is an area where arcuate edge series or intersections tend to gather outside the outer periphery of the color patch 1.

[Application example]
In the first embodiment, the case where the X direction filter and the Y direction filter are applied to the contour point included in each region and its surrounding pixels to calculate the joint angle of the contour point is illustrated. It is not necessary to apply the direction filter only to the contour points. For example, the disclosed apparatus applies various edge detection filters to each pixel included in the image to extract the edge from the luminance component of the image, and then calculates a connection angle for each contour point that constitutes the edge. It doesn't matter if you do.

  In the first embodiment, the case where the image processing apparatus 10 detects the measurement region formed by the through hole 3 of the color patch 1 is exemplified, but the application range of the disclosed apparatus is not limited to this. For example, the disclosed apparatus can detect an arbitrary object from an image even if it is an object other than the color patch 1 as long as the object is a concentric circle, ellipse, or polygon. However, it is not always necessary that a circle, an ellipse, or a polygonal multiplex be realized by the through holes.

[Distribution and integration]
In addition, each component of each illustrated apparatus does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, a part of the function units 15 a to 15 g in the control unit 15, for example, the association unit 15 f and the output unit 15 g may be connected as an external device of the image processing apparatus 10 via a network. In addition, the image processing apparatus 10 described above is configured by another device having a part of the function units 15a to 15g in the control unit 15, for example, the association unit 15f and the output unit 15g, which are connected to each other through a network. You may make it implement | achieve the function of.

[Image processing program]
The various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. In the following, an example of a computer that executes an image processing program having the same function as that of the above-described embodiment will be described with reference to FIG.

  FIG. 23 is a schematic diagram illustrating an example of a computer that executes an image processing program according to the first and second embodiments. As illustrated in FIG. 23, the computer 100 includes an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Further, the computer 100 includes a CPU 150, a ROM 160, an HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

  As shown in FIG. 23, the HDD 170 is similar to the acquisition unit 15a, grouping unit 15b, calculation unit 15c, extraction unit 15d, detection unit 15e, association unit 15f, and output unit 15g described in the first embodiment. An image processing program 170a that exhibits the function is stored in advance. The image processing program 170a is similar to the components of the acquisition unit 15a, grouping unit 15b, calculation unit 15c, extraction unit 15d, detection unit 15e, association unit 15f, and output unit 15g shown in FIG. You may integrate or isolate | separate suitably. In other words, all data stored in the HDD 170 need not always be stored in the HDD 170, and only data necessary for processing may be stored in the HDD 170.

  Then, the CPU 150 reads the image processing program 170 a from the HDD 170 and develops it in the RAM 180. Accordingly, as shown in FIG. 23, the image processing program 170a functions as an image processing process 180a. The image processing process 180a expands various data read from the HDD 170 in an area allocated to itself on the RAM 180 as appropriate, and executes various processes based on the expanded data. The image processing process 180a is performed by the acquisition unit 15a, grouping unit 15b, calculation unit 15c, extraction unit 15d, detection unit 15e, association unit 15f, and output unit 15g illustrated in FIG. 17 to 21 are included. In addition, each processing unit virtually realized on the CPU 150 does not always require that all processing units operate on the CPU 150, and only a processing unit necessary for the processing needs to be virtually realized.

  Note that the image processing program 170a is not necessarily stored in the HDD 170 or the ROM 160 from the beginning. For example, each program is stored in a “portable physical medium” such as a flexible disk inserted into the computer 100, so-called FD, CD-ROM, DVD disk, magneto-optical disk, or IC card. Then, the computer 100 may acquire and execute each program from these portable physical media. In addition, each program is stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, etc., and the computer 100 acquires and executes each program from these. It may be.

1 Color patch 3 Through hole C1, C2, C3, C4, C5, C6, C7, C8, C9, C10, C11, C12, C13, C14, C15, C16 Marking area W1, W2, W3, W4 White marking area DESCRIPTION OF SYMBOLS 10 Image processing apparatus 11 Camera 13 Memory | storage part 13a Connection angle data 13b Partial outline data 13c Voting data 15 Control part 15a Acquisition part 15b Grouping part 15c Calculation part 15d Extraction part 15e Detection part 15f Correlation part 15g Output part

Claims (7)

An acquisition unit for acquiring images;
A grouping unit that groups pixels having similar pixel values between adjacent pixels among the pixels included in the image, and divides the pixels into regions;
For each contour pixel that forms the contour of the grouped region, a calculation unit that calculates a connection angle at which the contour pixel is connected to another contour pixel on the contour of the region;
An extraction unit that extracts a partial contour line from the contour line by dividing the contour line forming the region by a section in which the change in the connection angle of the contour pixels is within a predetermined threshold;
An image processing apparatus comprising: a detection unit configured to detect a region having the largest number of intersections between the perpendicular lines of the partial contour lines from the region divided by the grouping unit. The detector is
Holds the number of intersections between the perpendiculars of the partial contour lines for each region,
For each combination of the partial contour lines, out of the number of intersections held for each region, a perpendicular line connecting the contour pixels at both ends of one partial contour line included in the combination and both ends of the other partial contour line Update the number of intersections in the area where the intersection with the perpendicular to the line segment connecting
The image processing apparatus according to claim 1, wherein an area having the largest number of intersection points is detected from areas divided by the grouping unit.   When the distance from the intersection point to the partial contour line is greater than the distance from the intersection point to the line segment connecting the contour pixels at both ends of the partial contour line, the detection unit calculates the number of intersection points of the region where the intersection point exists. The image processing apparatus according to claim 1, wherein the image processing apparatus is updated. The extraction unit includes:
Holds the articulation angle of the contour pixels that are candidates for the start point of the partial contour line,
Search for contour pixels on the contour line of the region,
Determining whether the difference between the concatenated angle of the searched contour pixel and the concatenated angle of the contour pixel held as a candidate for the start point of the partial contour line exceeds a predetermined division threshold;
When the difference exceeds the division threshold, a section from the contour pixel held as a candidate for the starting point of the partial contour to the contour pixel searched immediately before the contour pixel exceeding the division threshold is the partial contour. Extract as
The image processing apparatus according to claim 1, 2, or 3, wherein a concatenated angle of contour pixels exceeding the division threshold is overwritten on a concatenated angle of contour pixels held as a candidate for a starting point of a partial contour line. .   The calculation unit calculates a luminance gradient with a peripheral pixel of the contour pixel for each contour pixel, and calculates a direction perpendicular to a direction represented by the calculated luminance gradient as a connection direction of the contour pixels. The image processing apparatus according to claim 1. On the computer,
Get an image,
Grouping pixels that have similar pixel values between adjacent pixels among the pixels included in the image,
For each contour pixel that forms the contour of the grouped region, calculate the connection angle at which the contour pixel is connected to other contour pixels on the region contour,
Extracting a partial contour line from the contour line by dividing the contour line forming the region by a section in which the change in the connection angle of the contour pixels is within a predetermined threshold;
An image processing program for executing a process of detecting a region having the largest number of intersections between perpendiculars of the partial contour lines from the grouped regions. Computer
Get an image,
Grouping pixels that have similar pixel values between adjacent pixels among the pixels included in the image,
For each contour pixel that forms the contour of the grouped region, calculate the connection angle at which the contour pixel is connected to other contour pixels on the region contour,
Extracting a partial contour line from the contour line by dividing the contour line forming the region by a section in which the change in the connection angle of the contour pixels is within a predetermined threshold;
An image processing method comprising: executing a process of detecting an area having the largest number of intersections between perpendiculars of the partial contour lines from the grouped areas.