Color evaluation device and color evaluation method

Abstract

To provide a color evaluation device and a color evaluation method which can digitize a state that consecutively changes even when the entire measurement object is not in a uniform color and there is a certain degree of unevenness or a plurality of colors exist in a spotted manner.SOLUTION: A color evaluation device 100 comprises: an imaging unit 101 which acquires image data by imaging a measurement object; a reference color setting unit 104 which records setting information of a plurality of reference colors; an evaluation reference color designation unit 105 which designates one of the plurality of reference colors as an evaluation reference color on the basis of color information included in the image data; an approximate color region extraction unit 106 which extracts an approximate color region whose color approximates the evaluation reference color included in the measurement object of the image data; a ratio calculation unit 107 which calculates a ratio of the approximate color region in the measurement object; and an evaluation value calculation unit 109 which calculates an evaluation value for evaluating a color of the measurement object on the basis of the ratio of the approximate color region.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a color evaluation device that measures and evaluates the color of a product or the like using a color camera, and particularly to a color evaluation device that quantifies the state of food or the like whose surface color changes depending on the degree of heating.

Background Art

[0002] In Patent Document 1, in a food inspection device 11, the RGB color information of the image data of a good-quality bento box 2 captured by a color camera 4 is converted into HSL information, color information is extracted from the converted HSL information, the area of the ingredients is measured from the extracted color information, and ingredient omission is determined from the measured area. Note that RGB means the three primary colors of red, green, and blue, and HSL means the three components of hue, saturation, and lightness.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a demand to always maintain the processing state (such as heating state) of food or the like in a good (tasty) state and carry out production. If the processing time and the processing state of the product always have a fixed relationship, it is only necessary to manage the processing time to be always constant. However, in reality, it is necessary to adjust the processing time due to changes in the environment such as the origin and growth state of the raw material, the temperature and humidity on the processing day.

[0005] Therefore, in the past, the processing status of food products has been judged by their color in order to determine whether they have been overheated or underheated during processing. In addition to visual inspection, the color of the product may be quantified by photographing it with a color camera or multispectral camera, or by using measuring devices such as a spectrophotometer or colorimeter to quantify the color of the product.

[0006] However, when the object being measured is food, the entire product may not have a uniform color, and there is often some degree of unevenness or a patchy presence of multiple colors. Also, since it is often used to determine whether or not a particular sample color is present or to judge the difference from that sample color in order to identify defects in the object being measured, it is difficult to quantify continuously changing states such as the degree of heating of food.

[0007] Therefore, even when measurements are taken using high-precision spectrophotometers or colorimeters, the measured values ​​may be representative values ​​that represent the average value of only a part of the object being measured or a certain range, and there may be discrepancies with the actual situation.

[0008] Figure 17 illustrates the RGB and HLS measurements of snack foods (see Figure 8) with six different heating times. As shown in Figure 17, even if you cut out the central part of the snack food (the object T) and measure the RGB and HLS values ​​for the averaged color within that area, none of the measurements are suitable as indicators of the degree of heating. This is because in the range corresponding to the middle three stages of the six heating times (C3 to C5 in Figure 8), the measured values ​​are small or the correlation is reversed.

[0009] The purpose of this disclosure is to provide a color evaluation device and a color evaluation method that can quantify a continuously changing state, even when the entire object to be measured is not a uniform color, but has some degree of unevenness or multiple colors are present in a patchy manner. [Means for solving the problem]

[0010] The color evaluation apparatus disclosed herein comprises: an imaging unit that images a target to be measured and acquires image data; a reference color setting unit that records setting information for a plurality of reference colors; an evaluation reference color designation unit that designates one reference color from the plurality of reference colors as an evaluation reference color based on the color information contained in the image data; an approximate color region extraction unit that extracts approximate color regions in the image data whose color is similar to the evaluation reference color contained in the target to be measured; a ratio calculation unit that calculates the ratio of the approximate color regions to the target to be measured; and an evaluation value calculation unit that calculates an evaluation value for evaluating the color of the target to be measured based on the ratio of the approximate color regions.

[0011] The color evaluation method disclosed herein includes an imaging step of capturing an image of a measurement target to acquire image data; a reference color setting step of recording setting information for a plurality of reference colors; an evaluation reference color designation step of designating one reference color from the plurality of reference colors as an evaluation reference color based on the color information contained in the image data; an approximate color region extraction step of extracting approximate color regions in the image data whose color is similar to the evaluation reference color contained in the measurement target; a ratio calculation step of calculating the ratio of the approximate color regions to the measurement target; and an evaluation value calculation step of calculating an evaluation value for evaluating the color of the measurement target based on the ratio of the approximate color regions. [Effects of the Invention]

[0012] According to this disclosure, even if the entire object being measured does not have a uniform color, but rather has some degree of unevenness or multiple colors are present in a patchy manner, it becomes possible to quantify the continuously changing state. [Brief explanation of the drawing]

[0013] [Figure 1] This is a block diagram showing the configuration of a color measuring device 100 according to the first embodiment of the present invention. [Figure 2] This is a schematic diagram illustrating the actual equipment configuration of the color measuring device 100. [Figure 3] This figure shows a preferred example of the correspondence between the change in the area ratio of two reference colors and the evaluation value when the evaluation value is calculated using two reference colors. [Figure 4] This is a diagram showing an unfavorable example of the correspondence between the change in the area ratio of two colors and the evaluation value when an inappropriate reference color is selected in the case of calculating the evaluation value using two reference colors. [Figure 5] This is a diagram showing a favorable example of the correspondence between the change in the area ratio of three colors and the evaluation value in the case of calculating the evaluation value using three reference colors. [Figure 6] This is a diagram showing an unfavorable example of the correspondence between the change in the area ratio of three colors and the evaluation value when an inappropriate reference color is selected in the case of calculating the evaluation value using three reference colors. [Figure 7] This is a diagram showing an example of calculating the evaluation value from two reference colors. [Figure 8] This is a schematic diagram exemplifying the heating time of snack foods and the color change in six stages. [Figure 9] This is a schematic diagram showing three reference colors of the snack food illustrated in FIG. 8 and the area of the color extracted by each reference color. [Figure 10] This is a diagram comparing the areas respectively extracted from the snack food illustrated in FIG. 8. [Figure 11] This is an explanatory diagram exemplifying a method of calculating a linear evaluation value from the area ratio for each extracted color. [Figure 12] This is an explanatory diagram of the evaluation score value calculated by the method illustrated in FIG. 11. [Figure 13] This is a block diagram showing the configuration of the color measurement device 100A according to the second embodiment of the present invention. [Figure 14] This is a schematic diagram exemplifying the actual device configuration of the color measurement device 100A. [Figure 15] This is a block diagram showing the configuration of the color measurement device 100B according to the third embodiment of the present invention. [Figure 16] This is a block diagram showing the configuration of the color measurement device 100C according to the fourth embodiment of the present invention. [Figure 17] This is a diagram exemplifying the RGB and HLS measurement values of snack foods with different heating times in six stages.

MODE FOR CARRYING OUT THE INVENTION

[0014] <First Embodiment> 1.1 Configuration of Color Measurement Device 100 FIG. 1 is a block diagram showing the configuration of a color measurement device 100 according to the first embodiment of the present invention. FIG. 2 is a schematic diagram illustrating the actual device configuration of this color measurement device 100.

[0015] As shown in FIG. 1, the color measurement device 100 includes an imaging unit 101, an image buffer unit 102, a measurement position specifying unit 103, a reference color setting unit 104, an evaluation reference color specifying unit 105, an approximate color region extraction unit 106, a ratio calculation unit 107, an evaluation value interval setting unit 108, an evaluation value calculation unit 109, and an evaluation value output unit 110.

[0016] As an actual device configuration of the color measurement device 100, as shown in FIG. 2, it includes an imaging unit 101 and a PC (Personal Computer) 120.

[0017] The imaging unit 101 is arranged at a position suitable for imaging the measurement object T. The imaging unit 101 is, for example, a camera, and more specifically, a color camera capable of taking color images. It can be applied to both area sensor cameras and line sensor cameras, and can be combined with various lenses, filters, and even lighting devices according to the imaging object.

[0018] Elements other than the imaging unit 101 included in the color measurement device 100 are usually realized within a computer system, and in this color measurement device 100, they are configured as software inside the PC 120.

[0019] The image buffer unit 102 stores the image data generated by the camera in the imaging unit 101.

[0020] The measurement position identification unit 103 identifies the position of the object T being measured within the image being captured. This can be separated relatively easily by selecting materials for conveying equipment such as a conveyor belt that will serve as the background color, so that the background color and the color of the object T are sufficiently distinguishable in terms of brightness and hue. The position of the object T is then determined by the center point or centroid of the separated pixel set of the object T. Alternatively, the position can also be determined by searching within the image using known techniques such as pattern matching. In any case, determining the position of the object T within the image can be done using known techniques.

[0021] The reference color setting unit 104 allows the user to set multiple arbitrary colors as reference colors and as color range data. Specifying the color range by numerically representing the three elements of hue, saturation, and lightness (HSL) is preferable to specifying it using red, blue, and green (RGB), which is commonly used in computers, because it is closer to human perception and easier to understand.

[0022] The color range is defined by setting two values ​​for each of the three elements: hue, saturation, and lightness: a start value and an end value. However, it is important to note that for hue, since it loops in a circular fashion, the maximum and minimum values ​​of hue are adjacent colors. For example, if the possible values ​​for hue are integers from 0 to 359, the next value after 359 is 0. Therefore, it is possible to set the hue range as (start value ~ end value) = (359 ~ 1), and when extracting colors, the values ​​that fit the setting (359 ~ 1) are the three values ​​(359, 0, 1). For lightness and saturation, (start value ≤ end value) is always true.

[0023] The evaluation reference color specification unit 105 designates one reference color from among multiple reference colors as the evaluation reference color based on the color information contained in the image data stored in the image buffer unit 102.

[0024] The approximate color region extraction unit 106 extracts pixels of a color that approximates the evaluation reference color set by the reference color setting unit 104 based on the position information of the measurement target object T stored in the image data of the image buffer unit 102.

[0025] The ratio calculation unit 107 calculates the ratio of the number of pixels extracted by the approximate color area extraction unit 106 to the total number of pixels constituting the object T as a feature quantity. The total number of pixels constituting the object T is calculated when the object T is extracted by the measurement position identification unit 103.

[0026] The evaluation value interval setting unit 108 holds setting information for identifying which interval of the evaluation value a reference color feature (area ratio of reference colors) calculated using multiple reference colors falls into. For example, it can be set that reference color A is used to calculate evaluation values ​​from 0 to 30, and reference color B is used to calculate evaluation values ​​from 50 to 100. Alternatively, the evaluation values ​​from 30 to 50 may be calculated as the average of the values ​​obtained using reference color A and reference color B, or by using the larger of the reference color feature quantities of reference color A and reference color B. Rules for such calculations are set.

[0027] Even when there are three or more reference colors, you can arbitrarily divide the evaluation value and set reference color features to be used in each interval, or, if two reference colors are mixed, you can set which one to adopt or whether to adopt the average value.

[0028] The evaluation value calculation unit 109 calculates one evaluation value within the interval determined by the evaluation value interval setting unit 108.

[0029] The evaluation value output unit 110 displays one evaluation value calculated by the evaluation value calculation unit 109 on a monitor screen or saves it to a storage device such as external storage.

[0030] 1.2 About standard colors The reference color can be any color, preferably a distinctive color that clearly indicates the processing state of products such as food, but it is subject to the following constraints.

[0031] Figure 3 shows a preferred example of the correspondence between the change in the area ratio of two colors and the evaluation value when the evaluation value is calculated using two reference colors.

[0032] Graph C31 shows the relationship between the area ratio and the evaluation value when the color closest to the material is used as the reference color. Immediately after processing begins, when the material color occupies the entire area, the area ratio is almost 100%, and the evaluation value at this point is set to 100. As processing progresses, the material changes color, and the proportion of the material color decreases. Eventually, the area ratio becomes 0%, but the evaluation value at the point where it first becomes 0% is set to 25. The evaluation value at this point can be determined by human perception.

[0033] The relationship between the evaluation value and the area ratio can be arbitrarily determined based on the empirical relationship between processing time and discoloration. In this case, the interval in which graph C31 is changing (evaluation value: 25 to 100) can be approximately expressed by a linear function, i.e., the following linear equation.

[0034] If we let x be the area ratio and y be the evaluation value, then since it is a straight line passing through the two points (100,100) and (0,25), substituting these values ​​into the linear equation (y=ax+b), we get a=0.75 and b=25. That is, y = 0.75x + 25 It can be expressed as follows.

[0035] If we define the unprocessed state as having an evaluation value of 100 and the over-processed state (such as charred black) as 0, then one of the two colors must include the original material color before processing. In the unprocessed state with an evaluation value of 100, that color occupies almost 100% of the area of ​​the object T being measured. The graph of the reference color at this point is shown as C31. The closer the evaluation value is to 100, the closer the area ratio to the whole is to 100%, and as processing progresses and the evaluation value decreases, the area ratio also decreases.

[0036] The other color represents the color of an over-processed state, so it is desirable that it be a color indicating a burnt state, such as black or brown. However, there are cases where it is not necessary to calculate an evaluation value for a completely burnt state. In that case, you can choose a different color (the color of the state before it becomes burnt). Doing so will allow for a finer granularity of evaluation values ​​to be calculated for the state before and after the desired finished state.

[0037] The graph of the standard color under excessive processing is shown as C32. The closer the evaluation value is to 0, the closer the area ratio to the total is to 100%. In the interval where graph C32 changes (evaluation value: 0 to 60), it can be approximately represented by a linear function, i.e., the following linear equation.

[0038] If we let x be the area ratio and y be the evaluation value, then since it is a straight line passing through the two points (100,0) and (0,60), substituting these values ​​into the linear equation (y=ax+b), we get a=-0.6 and b=60. That is, y = -0.6x + 60 It can be expressed as follows.

[0039] Note that S31 is a section where it is preferable to calculate the evaluation value using C31, and S32 is a section where it is preferable to calculate the evaluation value using C32. P3 is the position where the area ratio of C31 and C32 is equal.

[0040] Figure 4 shows an undesirable example of the relationship between the change in the area ratio of two colors and the evaluation value when inappropriate reference colors are selected when calculating the evaluation value using two reference colors.

[0041] In Figure 4, although the graphs representing the area ratios of the two reference colors, shown as C41 and C42, change as the processing progresses, similar to Figure 3, the area ratios are all 0 in the processing stage shown as S43, making it impossible to calculate an evaluation value. In other words, since the evaluation value is the same, the changes in section S43 cannot be reflected in the evaluation value.

[0042] While such a selection of reference colors is undesirable, it is usually possible to avoid the occurrence of interval S43, where both area ratios are 0, by expanding the range of reference colors to increase the number of colors to be extracted.

[0043] However, simply broadening the range of reference colors can sometimes lead to ambiguity in evaluation values. In such cases, it is advisable to add a third reference color that appears between the transitions of the first and second colors.

[0044] Figure 5 shows a preferred example of the correspondence between the change in the area ratio of three reference colors and the evaluation value when the evaluation value is calculated using three reference colors.

[0045] Here, there is an interval S53 where the evaluation value cannot be calculated properly, similar to Figure 4, between the color C51, which is close to the original material, and the color C52, which is in a burnt state. However, by adding a third color, C53, it becomes possible to calculate the evaluation value even within the interval S53.

[0046] Unlike the other two colors, which show monotonically increasing or decreasing graphs, the third color graph exhibits both increases and decreases. Therefore, for example, two evaluation values, V51 and V52, are calculated for the same area ratio of 70%. However, within the interval S53, it shows a monotonically decreasing trend, and is determined solely by the evaluation value V52, so there is no problem in calculating the evaluation value.

[0047] Furthermore, the upper limit of the interval for which it is desirable to calculate the evaluation value using C53 is P51, and the lower limit is P52.

[0048] Figure 6 shows an undesirable example of the correspondence between the change in the area ratio of the three reference colors and the evaluation value when an inappropriate reference color is selected when calculating the evaluation value using three reference colors.

[0049] Depending on the selection of the reference color, the evaluation value may increase or decrease in section S63, where it cannot be calculated using the other two colors, such as C63. In such cases, it is not possible to eliminate section S63 where an evaluation value cannot be calculated. This is because, for example, two evaluation values, V61 and V62, are calculated for the same area ratio of 70%.

[0050] Possible solutions include slightly shifting the reference color to move the peak of the graph to the right or left, widening the range of the reference color C61 to adjust the point where the area ratio becomes 0 to the left, or widening the range of the reference color C62 to adjust the point where the area ratio becomes 0 to the right. However, it is highly likely that the selection of the third reference color, C63, is inappropriate, as it prevents accurate determination of the product's condition.

[0051] If that still doesn't work, we'll consider introducing a fourth color. The approach to the fourth color is the same as for the third color, so we'll omit the explanation.

[0052] Figure 7 shows an example of calculating an evaluation value from two reference colors.

[0053] For example, in Figure 7, when the area ratio of C31 extracted with the same settings as in Figure 3 is 50%, the aforementioned y = 0.75x + 25 The evaluation value can be calculated as 62.5 using this formula.

[0054] Furthermore, when the area ratio of C32 is 75%, as mentioned above... y = -0.6x + 60 The evaluation value can be calculated as 15 using this formula.

[0055] Based on the above, I will now explain some more specific examples.

[0056] 1.3 Specific Examples Figure 8 is a schematic diagram illustrating the heating time and color change of a snack food in six stages. Figure 9 is a schematic diagram showing the three reference colors of the snack food exemplified in Figure 8 and the area of ​​the color extracted by each reference color. Figure 10 is a diagram comparing the areas extracted from the snack food exemplified in Figure 8. Note that the snack food is an example of "food" in this disclosure.

[0057] As shown in Figure 8, the snack food changes color from C6 at the start of frying down to C1 as the heating progresses. The color change is not uniform throughout, but rather uneven, with some parts burning. Here, C1 to C6 represent the respective heating states as follows: C6: Deep-fried C5: Pure white C4: About standard C3: Slightly darker C2: Quite strong C1: Completely burnt

[0058] C6 still shows the natural color of the ingredients because the heating time is short. As heating time progresses, it gradually changes to reddish-brown. C5, C4, and C3 change to reddish-brown as heating progresses (from right to left in Figure 8). As heating time continues, it starts to burn, and the area of ​​black or dark brown increases. C1 indicates that it has been overheated and burnt to a crisp. Generally, C3 and C4 are considered to be the most delicious.

[0059] As shown in Figures 9 and 10, the color extracted using the reference color C91, which is close to the material's color (raw color), initially accounts for almost 100%, but its area decreases as the heating time progresses. The change in value is as shown for C51 in Figure 5.

[0060] The reference color C92 (red family) is set to closely match the color of the finished product after proper heating. The area ratio of the colors extracted using reference color C92 follows the trend shown in C53 of Figure 5.

[0061] The standard color C93 (burnt color scheme) is set to include a large amount of color representing burnt food caused by overheating. The area ratio of the colors extracted by standard color C93 follows the trend shown in C52 of Figure 5.

[0062] Figure 11 is an explanatory diagram illustrating a method for calculating a linear evaluation value from the area ratio of each extracted color. Figure 12 is an explanatory diagram illustrating the evaluation score value calculated using the method illustrated in Figure 11.

[0063] As shown in Figure 11, the area ratio and evaluation value graph for each color are used to define the region and formula to be used for the evaluation value of each color, as follows. Burnt type: F(x) = -25x + 25 (when x > 15% or y > 80%) Red system: F(y) = -60y + 85 (when y > 5% and x < 15%) Biosystem: F(z) = 52z + 55 (when z > 5% and y < 60%)

[0064] The formula for calculating a single evaluation value from these formulas is defined as follows: Evaluation score: F(t) = max(F(x), F(y), F(z))

[0065] Note that this calculation formula adjusts the score value for "(4) Average" to around 60, but it is not limited to this type of adjustment.

[0066] According to the calculation formula defined and adjusted in this way, a linear evaluation value can be obtained, as shown in Figure 12.

[0067] Furthermore, the color measuring device 100 according to the first embodiment described above corresponds to the color measuring method disclosed herein if each part shown in the block diagram of Figure 1 is implemented using software.

[0068] <Second Embodiment> Figure 13 is a block diagram showing the configuration of the color measuring device 100A according to the second embodiment of the present invention. Figure 14 is a schematic diagram illustrating the actual equipment configuration of this color measuring device 100A. The differences from the first embodiment will be explained below.

[0069] As shown in Figure 13, the color measuring device 100A further includes a device control unit 111 that controls the device using the evaluation value calculated by the evaluation value calculation unit 109 of the color measuring device 100 of the first embodiment.

[0070] Furthermore, as shown in Figure 14, the actual equipment configuration of the color measurement device 100A includes, in addition to the imaging unit 101 and PC 120, a control device 130 and a manufacturing device 140. The device control unit 111 may be configured as software within the PC 120, or as software within the control device 130.

[0071] This configuration allows the manufacturing device 140 to be controlled according to the evaluation values. For example, if the manufacturing device 140 is a heating device, the processing state of the product can be appropriately adjusted by performing feedback control such as adjusting the processing temperature, processing time, and conveying speed.

[0072] <Third Embodiment> Figure 15 is a block diagram showing the configuration of a color measuring device 100B according to the third embodiment of the present invention. The actual equipment configuration of this color measuring device 100B is the same as that of the first embodiment (see Figure 2). The differences from the first embodiment will be explained below.

[0073] As shown in Figure 15, the color measuring device 100B adds an interval average calculation unit 112 to the color measuring device 100 of the first embodiment. The evaluation value calculation unit 109 calculates the interval average value (moving average value) of the evaluation value and then passes the value to the evaluation value output unit 110. This makes it possible to suppress the effects of variation even for products where the reference color features vary greatly by smoothing out the changes in values ​​as an average value.

[0074] Since the interval average is calculated with the aim of reducing the variability of the reference color features and thus minimizing fluctuations in evaluation values, it may also be calculated using a low-pass filter calculation method that exhibits a similar effect.

[0075] <Fourth Embodiment> Figure 16 is a block diagram showing the configuration of the color measuring device 100C according to the fourth embodiment of the present invention. The actual equipment configuration of this color measuring device 100C is the same as that of the second embodiment (see Figure 14). The differences from the first to third embodiments will be explained below.

[0076] As shown in Figure 16, the color measuring device 100C of the fourth embodiment includes a device control unit 111 that controls the manufacturing device 140 in the same way as the device control unit 111 of the second embodiment, using evaluation values ​​calculated by the interval average calculation unit 112 of the color measuring device 100B of the third embodiment.

[0077] With this configuration, even for products with large variations in evaluation values, the manufacturing device 140 can be controlled according to the evaluation value calculated by the interval average. For example, if the manufacturing device 140 is a heating device, the processing state of the product can be appropriately adjusted by performing feedback control such as adjusting the processing temperature, processing time, and conveying speed.

[0078] Embodiments of this disclosure have been described above with reference to the drawings. However, this disclosure is not limited to the embodiments described above, and can be implemented in various forms without departing from its essence. Furthermore, various disclosures can be formed by appropriately combining the multiple components disclosed in the above embodiments. For example, some components may be removed from all the components shown in the embodiments. The drawings schematically show each component in order to make them easy to understand, and the number of each component shown may differ from the actual number due to the convenience of drawing creation. Also, each component shown in the above embodiments is just an example and is not particularly limiting, and various modifications are possible without substantially departing from the effects of this disclosure. [Industrial applicability]

[0079] This disclosure is applicable to the field of color evaluation devices that measure and evaluate color. [Explanation of Symbols]

[0080] 100 Color measurement device 100A Color Measurement Device 100B Color Measurement Device 100C Color Measurement Device 101 Imaging Unit 102 Image buffer section 103 Measurement position identification unit 104 Standard color setting section 105 Evaluation Criteria Color Specification Section 106 Approximate color region extraction part 107 Ratio Calculation Unit 108 Evaluation value interval setting section 109 Evaluation Value Calculation Unit 110 Evaluation Value Output Unit 111 Device Control Unit 112 Section Average Calculation Unit 120 PCs (Personal Computers) 130 Control device 140 Manufacturing equipment T Object to be measured

Claims

1. An imaging unit that captures an image of the object to be measured and acquires image data, A reference color setting unit that records setting information for multiple reference colors, An evaluation reference color designation unit that, based on the color information contained in the image data, designates multiple reference colors that are different from each other from among the multiple reference colors as multiple evaluation reference colors, A similar color region extraction unit extracts a plurality of similar color regions whose colors are similar for each of the evaluation reference colors included in the measurement target of the image data, A ratio calculation unit that calculates the ratio of each approximate color region to the measurement target, An evaluation value calculation unit calculates an evaluation value for evaluating the color of the object to be measured based on the ratio of each approximate color region, An evaluation value interval setting unit stores evaluation value interval setting information that divides the numerical range of the evaluation value calculated by the evaluation value calculation unit into a plurality of evaluation value intervals that correspond one-to-one with each of the plurality of reference colors, Equipped with, The color evaluation device includes an evaluation value calculation unit which calculates the evaluation value based on the ratio for each approximate color region and the evaluation value interval setting information stored in the evaluation value interval setting unit.

2. The color evaluation apparatus according to claim 1, wherein the evaluation reference color designation unit designates one of the plurality of evaluation reference colors as the evaluation reference color that maximizes the area in which the color approximates the reference color included in the measurement target of the image data.

3. The approximate color region extraction unit calculates a plurality of approximate pixel counts, which are the number of pixels whose colors are similar to each of the evaluation reference colors included in the measurement target of the image data. The color evaluation apparatus according to claim 1, wherein the evaluation value calculation unit calculates the evaluation value based on the ratio of each approximate number of pixels to the total number of pixels of the image data to be measured.

4. The system further includes an approximate color range setting unit that records setting information for the approximate color range for each of the aforementioned reference colors. The color evaluation apparatus according to claim 3, wherein the approximate color region extraction unit extracts regions included in the approximate color range for each evaluation reference color included in the measurement target of the image data as a plurality of approximate color regions.

5. The color evaluation device according to claim 1, wherein the evaluation value calculation unit calculates the evaluation value using a plurality of linear functions with the ratio for each approximate color region as a variable.

6. The color evaluation apparatus according to any one of claims 1 to 5, wherein the evaluation value interval setting information includes information of an evaluation function that calculates the evaluation value using the ratio of the approximate color area set for each evaluation value interval as a variable.

7. The color evaluation apparatus according to any one of claims 1 to 5, further comprising a device control unit that performs feedback control of at least one of the processing temperature of a heating device for heating the object to be measured, the processing time of the heating device, and the transport speed of the heating device based on the evaluation value calculated by the evaluation value calculation unit.

8. The object to be measured is food, A color evaluation apparatus according to any one of claims 1 to 5, which evaluates the color of the food during the heating process based on the evaluation value calculated by the evaluation value calculation unit.

9. The imaging step involves capturing an image of the object to be measured and acquiring image data, A reference color setting step that records setting information for multiple reference colors, A step of specifying evaluation reference colors, in which, based on the color information contained in the image data, multiple reference colors that are different from each other are designated as multiple evaluation reference colors from among the multiple reference colors, A step of extracting approximate color regions in which the colors are similar for each of the evaluation reference colors included in the measurement target of the image data, A ratio calculation step of calculating the ratio of each approximate color region to the measurement target, An evaluation value calculation step, which calculates an evaluation value for evaluating the color of the object to be measured based on the ratio of each approximate color region, A value interval setting step that stores evaluation value interval setting information which divides the numerical range of the evaluation value into a plurality of evaluation value intervals that correspond one-to-one with each of the plurality of reference colors, Includes, A color evaluation method comprising the step of calculating the evaluation value, which calculates the evaluation value based on the ratio for each approximate color region and the evaluation value interval setting information.

Images