JP5123509B2 - Measurement region setting device, measurement region setting method, and spectroscopic measurement device - Google Patents

Description

The present invention is based on FPD (Flat Panel Display) and PDP (Plasma).
The present invention relates to a measurement region setting device, a measurement region setting method, and a spectroscopic measurement device used at the time of measurement related to “color” such as spectral transmittance and spectral reflectance of a minute cell in a color filter or the like used in a display panel.

Conventionally, when measuring the “color” of an object to be measured, deciding where the measurement area should be is important for obtaining an accurate measurement value.
As a method for determining a measurement region, for example, methods described in Patent Document 1 and Patent Document 2 are known.

The method of Patent Document 1 is a method for determining a measurement region of a three-dimensional image obtained from a color confocal microscope, and a region having the same information as information on “color” designated by an operator is used as a measurement region. Has been decided.
Moreover, the method of patent document 2 installs a translucent film in the part to measure in the whole area | region used as a measuring object as an area | region when measuring the "color" of the measuring object to which metallic coating was given. This makes it possible to accurately measure only the color on the light-transmitting film.
JP 2003167201 A JP-A-8-219885

However, with the methods disclosed in these Patent Documents 1 and 2, it is difficult to automatically set the measurement region.
In particular, if the area that has the “color” to be measured has a complex shape, the measurement area may include areas having other colors, and accurate measurement results cannot be obtained. There is an inconvenience.

  The present invention has been made in view of the above-described circumstances, and even when the color region to be measured has a complicated shape, the measurement region is accurately set so as not to include other colors. An object of the present invention is to provide a measurement region setting device, a measurement region setting method, and a spectroscopic measurement device.

In order to achieve the above object, the present invention provides the following means.
The present invention provides a color area extraction unit that extracts a color area having a specified color from a color image to be measured, and a color area feature extraction that extracts a feature amount of the color area extracted by the color area extraction unit. A search area extraction unit for extracting a search area for determining a measurement area from the color area based on the feature amount extracted by the color area feature extraction unit, and a search extracted by the search area extraction unit have a measurement region determination unit for determining a measuring area from the region, the measurement region determining unit divides a plurality of information calculation area on the basis of the previously set measurement area size the search area, designated measurements to calculate the remaining standard deviation for the specified color for each information calculation region, determines the information operation area in which the standard deviation is minimized as the measurement region excluding the information operation area which does not include the color To provide a frequency setting device.

In the above invention, the feature amount extracted by the color region feature extraction unit includes one or more of the center of gravity of the color region, the position of an arbitrary point on the circumscribed rectangle of the color region, and the area of the color region. Also good.
In the invention described above, the search area may be an area where the distance between the center of the color image and a center of the color area or an arbitrary point on a circumscribed rectangle is minimized.

  In the invention described above, the search area may be an area where the area of the color area is maximized.

Moreover, in the said invention, the said color area extraction part is good also as extracting the said color area based on at least hue information among the information contained in the said color image.
In the above invention, the color information extraction unit divides the color image into color-specific images represented by colors, selects the color-specific images required based on the specified colors, and selects The color area may be extracted based on the luminance of the color-specific image.

  Further, in the above invention, a relative measurement position storage unit that stores relative position information of the measurement region in the search region is provided, and the measurement region determination unit is stored in the relative measurement position storage unit. The measurement area may be determined using the relative position information of the position.

  In the above invention, the relative measurement position storage unit stores area information of the search region, and the measurement region determination unit includes the previous area information stored in the relative measurement position storage unit and the current area information. When the difference from the area information is equal to or less than a predetermined value, the measurement region may be determined using the relative position information.

In the above invention, the area information may be at least one of the area of the search region, the width and the height of the search region.
In the above invention, the relative measurement position storage unit stores image information of the search area, and the measurement area determination unit stores the previous image information stored in the relative measurement position storage unit and the current image information. Pattern matching processing with image information is performed, and when the evaluation value indicating the degree of matching of pattern matching is a predetermined value or more, the measurement area may be determined using the relative position information.

The present invention is also a method for setting a measurement region by image processing of a color image to be measured, wherein a color region indicating a specified color is extracted from the color image, and the extracted feature value of the color region And extracting a search area for determining a measurement area from the color area based on the extracted feature amount, and dividing the search area into a plurality of information calculation areas based on a predetermined measurement area size. Calculating the standard deviation for the designated color for each remaining information computation area excluding the information computation area not including the designated color, and determining the information computation area that minimizes the standard deviation as the measurement area A measurement area setting method is provided.

  In addition, the present invention provides an image acquisition unit that acquires a color image of a measurement target, any one of the measurement region setting devices described above, and spectral transmittance or reflectance in the measurement region calculated by the measurement region setting device. A spectroscopic measurement device including a spectroscopic measurement unit for calculation is provided.

  According to the present invention, even when the color region to be measured has a complicated shape, the measurement region can be accurately set so as not to include other colors.

A measurement region setting device 1 and a spectroscopic measurement device 2 according to a first embodiment of the present invention will be described with reference to FIGS.
A measurement region setting apparatus 1 according to the present embodiment is provided in a spectroscopic measurement apparatus 2 shown in FIGS.

The spectroscopic measurement apparatus 2 according to this embodiment includes a stage 3 on which a color filter A (a measurement target) is mounted, and a portal frame 5 that is linearly movable in the X direction along a rail 4 provided on the stage 3. A transmission illumination system 6 that is linearly movable in the Y direction below the color filter A along the rail 11 fixed to the portal frame 5 and a rail (not shown) fixed to the portal frame 5. A microscope head 7 that can move linearly in the Y direction above the color filter A, and a measurement region setting device that sets a measurement region by processing a color image acquired by the imaging device 8 provided in the microscope head 7 ( An image processing unit 1, a displacement amount detection unit 9 that detects a displacement amount between the color filter A and the optical axis of the microscope head 7 based on the measurement region set by the measurement region setting device 1, and the displacement amount detection Based on the displacement amount detected by the section 9, and a drive control unit 10 for moving the portal frame 5, a transmission illumination system 6 and the microscope head 7.
The spectroscopic measurement device 2 includes a spectroscopic calculation device (not shown) that calculates the spectral transmittance of the measurement region set by the measurement region setting device 1.

  The transmitted illumination system 6 is attached to a follow-up movement mechanism 13 provided on a slider 12 that can move along a rail 11 fixed to the portal frame 5. The following movement mechanism 13 can move the transmitted illumination system 6 following the movement of the microscope head 7 in the Y direction above the color filter A.

  On the slider 12, when the microscope head 7 is positioned at a position corresponding to one pixel (measurement region) of the color filter A, the measurement region and the transmission illumination system are based on the image data acquired by the imaging device 8. An XY table 14 for correcting a positional deviation from the spot illumination light due to 6 is provided. That is, the XY table 14 is driven by the operation of the drive control unit 10 so that the optical axis of the transmission illumination system 6 coincides with the optical axis of the objective lens 15 provided in the microscope head 7. The transmitted illumination system 6 is moved.

  As shown in FIG. 2, the measurement area setting device 1 includes color image information aa sent from the imaging device 8, user-specified measurement color information bb input from the spectroscopic calculation device, and the size of the measurement region. Is a device that receives measurement region size information cc representing the output and outputs measurement region position information hh, and includes a preprocessing unit 19, a measurement color region extraction unit 20, a measurement color region feature extraction unit 21, and a measurement color search. An area extraction unit 22 and a measurement area determination unit 23 are provided.

The preprocessing unit 19 receives and stores the color image information aa from the imaging device 8, and performs correction processing such as shading correction and distortion correction as necessary.
The measurement color region extraction unit 20 performs measurement among the color image information dd based on the color image information dd after correction processing input from the preprocessing unit 19 and the measurement color information bb input from the spectroscopic calculation device. An area having color information that substantially matches the color information bb is extracted.

  The measurement color information bb is a hue Hu, saturation Sa, and luminance Lu obtained from the color image information aa by HSL conversion. Hue Hu is expressed as an angle (0 ° ≦ Hu <360 °), saturation Sa is expressed as a percentage (0% ≦ Sa ≦ 100%), and luminance Lu is expressed as an image gradation (0 ≦ Lu ≦ 255).

The measurement color region feature extraction unit 21 receives image information ee of the measurement color region from the measurement color region extraction unit 20, and information on the measurement color region includes the area of the measurement color region, the barycentric position coordinate, circumscribed rectangular coordinates, and the like. The feature information ff is calculated and output.
The measurement color search region extraction unit 22 is configured to measure the measurement region based on the measurement color region image information ee output from the measurement color region extraction unit 20 and the feature information ff output from the measurement color region feature extraction unit 21. A search area for searching the position information hh is extracted, and search area information gg is output.

  The measurement area determination unit 23 receives image information ee indicating the measurement color area, search area information gg, and measurement area size information cc, and the image information located on the search area information gg is the same size as the measurement area size information cc. The position of the measurement area position information hh to be arranged is determined and output.

The operation of the measurement region setting apparatus 1 according to the present embodiment configured as described above will be described.
First, as shown in FIG. 3, the pre-processing unit 19 pre-processes the color image information aa input from the imaging device 8 (S101), and the measurement color relating to the color designated by the measurement color region extraction unit 20 A region is extracted (S102).

  As shown in FIG. 4, the color area is extracted by first setting the hue Hu as the threshold value Th_Hu1 included in the user-specified measurement color information bb with respect to the color image information dd input from the preprocessing unit 19. An area existing between the threshold value Th_Hu2 is extracted to generate first partial image information (S111). For example, if the hue Hu is based on green (Hu = 0 °: counterclockwise is the positive direction), if the input color information is red, the hue Hu indicating 240 ° is 240 °. The threshold values Th_Hu1 = 225 ° and Th_Hu2 = 255 ° are set in a range of ± 15 ° centered on the red hue.

  Next, with respect to the first partial image information extracted based on the hue threshold value, a region where the luminance Lu exists between the threshold value Th_Lu1 and the threshold value Th_Lu2 included in the measurement color information is extracted and the second partial image information is extracted. The partial image information is generated (S112). For example, Th_Lu1 = 60 (dark side) and Th_Lu2 = 190 (bright side) are set. This is because a portion that is too dark or too light is deleted and only a region where a more accurate color is reproduced is extracted.

  Then, with respect to the second partial image information extracted by the luminance threshold value, an area where the saturation Sa is equal to or higher than the threshold value Th_Sa included in the measurement color information is extracted to generate a measurement color area (S113). . For example, by setting the threshold Th_Sa = 30%, an area where the saturation is low, that is, an area where the color is close to an achromatic color is deleted, and a more vivid color is reproduced is used as the measurement color area image information ee. It can be extracted.

Next, when the extracted measurement color region image information ee is input to the measurement color region feature extraction unit 21 and the plurality of measurement color region image information ee is included in the color image information aa, all of the measurement color region image information ee is included. For the measurement color region image information ee, feature information ff such as an area, a gravity center position, a circumscribed rectangle position, and the like is calculated (S103).
Then, the measurement color search region extraction unit 22 performs a search based on the measurement color region image information ee output from the measurement color region extraction unit 20 and the feature information ff output from the measurement color region feature extraction unit 21. Area information gg is generated.

  Specifically, first, a measurement color region whose area in each measurement color region image information ee is equal to or less than a threshold value Th_Me1 is excluded (S104). Here, the threshold value Th_Me1 is set to about 1 to 5% with respect to the area of the entire color image, for example. A color area having a small area is excluded because accurate color measurement is impossible as a result.

  Next, for the measurement color region image information ee remaining without being excluded in step S104, the distance between the center of gravity of each region and the center position of the color image is calculated, and regions that are equal to or greater than the threshold Th_Me2 are excluded. (S105). Here, the threshold value Th_Me2 is set to about 20% of the diagonal distance of the color image information aa, for example. As the distance from the image center increases, the effect of image distortion and shading increases even if correction processing is performed, and it is desirable to minimize the distance traveled to the actual measurement position. The near measurement color region image information ee is selected.

Further, the distance between each position on the circumscribed rectangle of each area and the center of the color image is calculated for the measurement color area image information ee that is not excluded in step S105, and the calculated distance is the minimum. An area is extracted as search area information gg (S106).
As each position on the circumscribed rectangle, there are four corner points of the circumscribed rectangle. Among these, the regions having the smallest distance from the center of the color image information aa are compared between the regions.
Thereby, it is possible to extract the search area information gg that is close to the center of the color image information aa and whose area is not too small.

More specifically, as shown in FIG. 5, the search area ggg is extracted from the color image aaa shown in FIG. 5A acquired by the operation of the imaging device 8.
That is, when red is used as the measurement color, first, the measurement color region extraction unit 20 generates an image G1 including the four red measurement color regions eeee1 to eeee4 as shown in FIG.

  Here, the rightmost area in FIG. 5B is originally the same area, but is recognized as a separate area on the image. Then, the measurement color regions eee1 to ee4 extracted in this way are input to the measurement color region feature extraction unit 21, whereby the respective areas are calculated and as shown in FIG. 5C. The feature information ff such as the gravity center positions C1 to C4 and the circumscribed rectangles S1 to S4 is calculated.

  Furthermore, based on the calculated feature information ff and measurement color region image information ee, the operation of the measurement color search region extraction unit 22 first excludes a color region whose area is smaller than a predetermined value, and then the gravity center position C1. Measurements in which the distance between C4 and the center position of the color image aaa is larger than a predetermined value are excluded, and the distance between the four corner points on the circumscribed rectangles S1 to S4 and the center position of the color image aaa is minimized. As shown in FIG. 5D, an image G2 is obtained in which the color area eeee2 is extracted as the search area information ggg.

Then, the search area information gg, the measurement color area image information ee, and the measurement area size information cc thus extracted are input to the measurement area determination unit 23, and the size (width, height) of the circumscribed rectangle S2 and the measurement area are measured. Candidates for determining the position of the measurement region position information hh are defined using the size (width, height) of (S107).
Specifically, as shown in FIG. 6A, when the measurement color region eee2 has a complicated shape, a circular measurement region hh that performs actual measurement is located at the center of gravity of the search region ggg. When −0 is arranged, a different color region or background color is included in the measurement region hh-0, and accurate measurement cannot be performed.

  Therefore, as shown in FIG. 6B, the search area ggg has the same size as the circular measurement area hh-0 (the width and height are the same as the diameter dimension of the measurement area hh-0 having a circular shape). ) Measurement candidate areas (information calculation areas) hh′-1 to hh′-4. The measurement candidate regions hh ′ may be defined so as to overlap each other. By doing so, it is possible to define a measurement candidate region hh ′ that is slightly shifted, and it is possible to define a measurement candidate region hh ′ that includes only the color to be measured. FIG. 6 shows only some measurement candidate regions.

  Then, for the plurality of defined measurement candidate areas hh ′, only those that completely overlap with the measurement color area image information eee 2 are selected. As a result, as shown in FIG. 6C, the measurement candidate region hh′-2 in FIG. 6B is excluded.

  Thereafter, the standard deviation relating to the measurement color is calculated for the remaining measurement region candidates hh′-1, hh′-3, hh′-4, etc. (S108). Specifically, the calculated standard deviation is a standard deviation regarding the hue Hu. Then, by extracting the measurement region candidate hh′-1 that minimizes the calculated standard deviation and calculating the center position thereof, the position of the measurement region hhh can be finally determined (S109).

  Thus, according to the measurement area setting device 1 according to the present embodiment, the measurement color area image information ee including the specified color is extracted from the acquired color image information aa, and based on the feature information ff, Since the search area information gg is narrowed down, it is possible to extract search area information gg that is close to the center of the color image information aa and is not too small. Also, measurement area candidates that do not include the designated color are excluded from the search area, and an area with the smallest standard deviation regarding the color is set in the measurement area position information hh. Can be suppressed. Then, according to the spectroscopic measurement device 2 according to the present embodiment including the measurement region setting device 1, it is possible to perform accurate measurement while suppressing the influence of variations in measurement.

  In the present embodiment, the measurement color search region extraction unit 22 extracts the search region information gg so as to narrow down using different feature information ff one by one as in steps S104 to S106. Instead, the extraction of the search area information gg is not limited to this. For example, the measurement color region image information ee maximizing the area by focusing only on the area may be extracted as the search region information gg as it is, or the measurement having the centroid position where the distance from the center of the color image information aa is minimum. The color area image information ee may be extracted as the search area information gg.

  In the present embodiment, the measurement color region image information ee is extracted from information obtained by HSL conversion of the color image information aa. Instead, the color image information aa itself is used. May be. For example, when the color image information aa is composed of RGB components and the red region is extracted, the luminance information of the red component is used as it is and the luminance information is in a predetermined range (luminance information). May be extracted as measurement color region image information ee for determining the measurement region position information hh.

  Further, the measurement color region image information ee may be extracted by controlling other than the measurement region setting device 1 according to the color of the color filter A. For example, the exposure time of the imaging device 8 is changed according to the color of the color filter A, and the measurement area setting device 1 receives the exposure time as the measurement color information bb of the user instruction, and specifies the color according to this value. Also good. For example, when measuring green with a large luminance component, the exposure time is relatively short, and when measuring blue with a small luminance component, the exposure time is relatively long. This can be realized by preparing a table storing the relationship with the color to be measured.

  Moreover, you may make it change the light quantity of the illumination which injects into the imaging device 8 according to the color of a measuring object. Further, when the spectrum of the illumination light captured by the imaging device 8 affects the extraction of the measurement color region image information ee, according to the color to be measured by a bandpass filter that transmits only light having a wavelength on the illumination light side. It is good also as preparing two or more.

Next, a measurement region setting device 30 according to a second embodiment of the present invention will be described with reference to FIGS.
In the description of the present embodiment, portions having the same configuration as those of the measurement region setting apparatus 1 according to the first embodiment described above are denoted by the same reference numerals and description thereof is omitted.

As shown in FIG. 7, the measurement region setting device 30 according to the present embodiment is different from the measurement region setting device 1 according to the first embodiment in that it includes a relative measurement position storage unit 31.
The relative measurement position storage unit 31 receives the determined measurement region information hh and the search region information gg used by the measurement region determination unit 23, and calculates and stores the relative position of the measurement region with respect to the search region. It has become.

More specifically, as shown in FIG. 8, the search area information ggg2 including the circumscribed rectangle S indicated by a broken line and the circular measurement area position information hhh2 are set for the red measurement color area. Illustrate. Here, the coordinate values in FIG. 8 are image coordinate values, and the upper left corner of the color image information aaa2 is the origin, the rightward direction indicates the X direction, and the downward direction indicates the positive direction. At this time, the coordinates of the upper left corner of the search area information ggg2 are stored as (SPos-X1, SPos-Y1), and the center of gravity position of the determined measurement area position information hh2 is stored as (Pos-X, Pos-Y). The relative position information kk (RPos-X, RPos-Y) of the measurement region position information hh is calculated by the following equation.
RPos-X = Pos-X-SPos-X1
RPos-Y = Pos-Y-SPos-Y1

  In addition, the relative measurement position storage unit 31 stores the area of the search area and the size (width, height) of the circumscribed rectangle S from the search area information gg as area information of the search area. The area of the search area can be calculated in units of the number of pixels in the search area, and the size of the circumscribed rectangle can be calculated in units of pixels based on the upper left corner coordinates and the lower right corner coordinates of the circumscribed rectangle S in FIG.

In determining the measurement region, the relative measurement position storage unit 31 receives the search region information gg input from the measurement color search region extraction unit 22, and determines the relative position of the stored measurement region position information hh as the measurement region determination unit. 23, it is determined whether it can be used.
In this determination method, first, the search area information gg extracted in the current measurement is read (S121), and the area A0 of the search area information gg and the size of the circumscribed rectangle (width W0, height H0) are calculated (S122). ).

  Next, the area information (area Ab of the search area, size of the circumscribed rectangle (width Wb, height Hb)) stored in the previous process is read (S123). Then, as a comparison between the current search area and the previous search area, the areas of both are compared. Specifically, it is determined whether or not the absolute difference between the two areas is less than a threshold value Th_Me3 relating to the area (S124). The threshold value Th_Me3 is set to a small value, for example, about 1% to 5% as a ratio to the area of the color image information aa.

  If the absolute difference in area is less than the threshold Th_Me3, whether the absolute difference in width is less than the threshold Th_Me4 and the absolute difference in height is less than Th_Me5 with respect to the absolute difference in size of the circumscribed rectangle S Is determined (S125). The threshold values Th_Me4 and Th_Me5 are set to be small, for example, about 1% to 3% as a ratio to the width and height of the color image information aa.

  When the absolute difference in area, the absolute difference in width, and the absolute difference in height are all less than the threshold value, it can be determined that the search area for the stored measurement area and the current search area are substantially identical in shape. . Therefore, in this case, in order to perform the process of determining the measurement region position information hh using the stored relative position, the measurement region determination unit 23 stores the relative position information kk of the measurement region from the relative measurement position storage unit 31. Is output (S126).

  On the other hand, if either the area or the size of the circumscribed rectangle is greater than or equal to the threshold value, it is determined that the shape of the stored search area and the current search area are not the same. Therefore, the measurement area determination unit 23 outputs NULL information as the relative position information kk from the relative measurement position storage unit 31 in order to perform the processing shown in the first embodiment (S127).

  The measurement region determination unit 23 determines a method for determining the measurement region position information hh according to the input relative position information kk. That is, if the relative position information kk is NULL information, the measurement region position information hh is determined by the processing shown in the first embodiment. On the other hand, when the relative position information kk is not NULL information, the measurement area position information hh is determined by the following equation.

Pos-X '= SPos-X1' + RPos-X
Pos−Y ′ = SPos−Y1 ′ + RPos−Y
Here, (SPos-X1 ′, SPos-Y1 ′) are the coordinates of the upper left corner of the circumscribed rectangle relating to the search area extracted from the current color filter A, and (RPos-X, RPos-Y) are relative measurement position storage units. The relative position information kk is stored in 31.
Then, the measurement region position information hh output from the measurement region determination unit 23 is sent to the deviation amount detection unit 9 and the relative measurement position storage unit 31.

  FIG. 8B shows a case where the relative position information kk of the measurement area position information hh is used. FIG. 8C shows a case where the relative position information is not used. FIG. 8A shows the previous measurement area hh2 and search area ggg2 in which the relative position information kk is stored in the relative measurement position storage unit 31, and FIGS. 8B and 8C show the current measurement area hh3. , Hhh4 and search regions ggg3 and ggg4.

  In the example of FIG. 8B, the state is shifted to the right with respect to FIG. 8A, and there is no difference in the shape of the red measurement color region. Therefore, the position of the measurement region hh3 can be determined from the coordinates of the upper left corner of the circumscribed rectangle S in FIG. 8B using the relative position information kk output from the relative measurement position storage unit 31. Thereby, it is not necessary to perform the measurement area determination process in the first embodiment, and the processing time can be saved.

  On the other hand, in the example of FIG. 8C, the state is shifted downward with respect to FIG. As for the shape of the red measurement color region, there is a difference in the upper part of the image between them, and the width of the circumscribed rectangle S is thereby different. For this reason, NULL information is output from the relative measurement position storage unit 31, and for the circumscribed rectangle of the measurement color area in FIG. 8C, the position of the measurement area is determined by the method described in the first embodiment. .

  In the present embodiment, the operation of the measurement region determination unit 23 is made different based on the area information of the search region information gg. However, instead of this, the image information itself may be used. For example, instead of the area information, image information of the search area may be stored, and the image information may be compared by pattern matching processing when new image information of the search area is obtained. As a result, if the degree of coincidence is greater than a predetermined threshold value, the position of the measurement region can be calculated based on the relative position information kk of the measurement region position information hh stored in the relative measurement position storage unit 31. Such a method is effective in the case of measurement color regions having different shapes even if the area and circumscribed rectangle are similar.

1 is an overall configuration diagram showing a spectrometer according to an embodiment of the present invention. It is a block diagram which shows the measurement area | region setting apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the measurement area | region setting apparatus of FIG. 3 is a flowchart showing an operation of extracting a measurement color region of the measurement region setting apparatus in FIG. It is explanatory drawing which shows the extraction operation | movement of the search area | region by the measurement area | region setting apparatus of FIG. It is explanatory drawing which shows the determination operation | movement of the measurement area | region in the search area | region extracted in FIG. It is a flowchart which shows the measurement area | region setting apparatus which concerns on the 2nd Embodiment of this invention. It is explanatory drawing which shows the determination operation | movement of the measurement area | region in the measurement area | region setting apparatus of FIG. It is a flowchart which shows utilization of relative position information in determination operation | movement of the measurement area | region of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Measurement area setting apparatus 2 Spectrometer 20 Measurement color area extraction part (color area extraction part)
21 Measurement color region feature extraction unit (color region feature extraction unit)
22 Measurement color search area extraction unit (search area extraction unit)
31 Relative measurement position storage section aa Color image information ee Measurement color area image information (color area)
ff Feature information gg Search area information hh Measurement area position information A Color filter (measurement target)
C1-C4 center of gravity S1-S4, S circumscribed rectangle

Claims (12)

A color region extraction unit that extracts a color region having a specified color from the color image of the measurement target;
A color region feature extraction unit that extracts a feature amount of the color region extracted by the color region extraction unit;
A search region extraction unit that extracts a search region for determining a measurement region from the color region based on the feature amount extracted by the color region feature extraction unit;
A measurement region determination unit that determines a measurement region from the search regions extracted by the search region extraction unit ;
The measurement area determination unit divides the search area into a plurality of information calculation areas based on a predetermined measurement area size, and regarding each remaining information calculation area excluding an information calculation area that does not include a specified color A measurement area setting device that calculates a standard deviation for the designated color and determines the information calculation area that minimizes the standard deviation as the measurement area.   The measurement according to claim 1, wherein the feature amount extracted by the color region feature extraction unit includes one or more of a center of gravity of the color region, a position of an arbitrary point on a circumscribed rectangle of the color region, and an area of the color region. Area setting device.   The measurement area setting apparatus according to claim 2, wherein the search area is an area where a distance between a center of the color image and an arbitrary point on a center of gravity or a circumscribed rectangle of the color area is minimized.   The measurement area setting device according to claim 1, wherein the search area is an area in which an area of the color area is maximized.   The measurement area setting device according to claim 1, wherein the color area extraction unit extracts the color area based on at least hue information among information included in the color image. The color area extraction unit divides the color image into color-specific images represented by colors, selects the color-specific images required based on the designated colors, and selects the selected color-specific images. The measurement area setting device according to claim 1, wherein the color area is extracted based on luminance. A relative measurement position storage unit that stores relative position information of the measurement area in the search area;
The measurement region setting device according to claim 1, wherein the measurement region determination unit determines a measurement region using relative position information of the measurement position stored in the relative measurement position storage unit. The relative measurement position storage unit stores area information of the search region,
The measurement region determination unit determines a measurement region using the relative position information when a difference between the previous area information stored in the relative measurement position storage unit and the current area information is equal to or less than a predetermined value. The measurement region setting device according to claim 7.   The measurement area setting device according to claim 8, wherein the area information is at least one of an area of the search area, a width and a height of the search area. The relative measurement position storage unit stores image information of the search area;
The measurement area determination unit performs a pattern matching process between the previous image information stored in the relative measurement position storage unit and the current image information, and an evaluation value indicating a degree of coincidence of pattern matching is a predetermined value or more. The measurement area setting device according to claim 7, wherein the measurement area is determined using the relative position information. A method of setting a measurement region by image processing of a color image to be measured,
Extract the color area indicating the specified color from the color image,
Extracting the feature amount of the extracted color region;
Based on the extracted feature quantity, extract a search area for determining a measurement area from the color area,
The search area is divided into a plurality of information calculation areas based on a predetermined measurement area size, and the standard deviation relating to the specified color with respect to each remaining information calculation area excluding the information calculation area not including the specified color , And the information calculation area that minimizes the standard deviation is determined as the measurement area. An image acquisition unit for acquiring a color image of the measurement target;
The measurement region setting device according to any one of claims 1 to 10,
A spectroscopic measurement device comprising: a spectroscopic measurement unit that calculates spectral transmittance or spectral reflectance in a measurement region set by the measurement region setting device.

Images