TWI695169B - Color comparison area selection method - Google Patents

Abstract

Color comparison area selection method is disclosed. First, the color difference between the plurality of color comparison regions and the sample region is analyzed and sorted by using different color analysis methods to form color comparison region orders corresponding to the color analysis methods respectively. Next, the score of each color analysis method is calculated, and at least two color analysis methods which the scores are closest are selected to set the first weight. The other color analysis methods are set other weights smaller than the first weight. Then, the score of each comparison region is calculated according to the first weight and the other weights. The color of the comparison region with the lowest score is selected to be as the color of the sample region

Description

Selection method of color comparison area

The present invention relates to a selection method, in particular to a method for selecting a color comparison area.

Generally, when performing health tests, blood, urine or other body fluids taken from the human body are often used in conjunction with corresponding test papers for in vitro testing to provide real-time color display data for medical personnel to judge the health status of the testee.

When the tester uses the test paper to perform the test, the color analysis method is usually used to confirm the color on the test paper and which color of the comparison area is the most similar, and then determine the test result. However, different analysis methods may have different interpretation results for the same color, so that the test results of the testee often have different conclusions due to the different color analysis methods.

Therefore, how to choose the correct analysis method to obtain the correct color judgment result and then improve the detection accuracy has become the goal of pursuit.

An embodiment of the present case provides a method for selecting a color comparison region, which is used to select a comparison region from the plurality of color comparison regions that is closest to the color presented by the same region. This method first uses a plurality of color analysis methods to analyze the color difference between the plurality of color comparison regions and the sample region according to the same region, and sorts them to form a plurality of color comparison regions, which respectively correspond to the plurality of color analysis methods. The sample area is used to receive a subject's body fluid and correspondingly display a color. Then, the score of each of the plurality of color analysis methods is calculated, and at least two color analysis methods with the closest scores are selected from the color analysis methods. Next, set the at least two color analysis methods to have a first weight, and set the remaining color analysis methods to have other weights different from the first weight, where the first weight is greater than the other weights, and according to the first The weight and the other weights are used to calculate the score of each comparison area. Then compare the scores of each comparison area to select the lowest score from the comparison area that is closest to the color presented by the sample area.

According to an embodiment of the present invention, the sample area is used to receive urine from a subject.

According to an embodiment of the present invention, wherein the score of each of the plurality of color analysis methods is calculated using a discounted cumulative gain method (Discounted Cumulative Gain, DCG), where:

Among them, K represents the k-th color comparison region; Color_system_score _{i, k} represents the score of the _i- th color analysis method in the k-th color comparison region.

According to an embodiment of the present invention, the score of the i-th color analysis method in the k-th color comparison region is the color comparison of the k-th color comparison region in the i-th color analysis method Sorting position in area sorting.

According to an embodiment of the present invention, wherein calculating the score of each comparison area according to the first weight and the other weights further comprises: dividing the first weights respectively Multiplying each of the comparison area scores in the at least two color analysis methods to obtain a first weight type score; multiplying the other weights by the corresponding each of the comparison area scores in the other color analysis methods respectively Other weight type scores; and the first weight type score and the other weight type scores are added together as the score of each comparison area.

According to an embodiment of the present invention, wherein the score of each comparison region in the at least two color analysis methods is the ranking of each comparison region in the color comparison region corresponding to the at least two color analysis methods Sort position.

According to an embodiment of the present invention, the score of each comparison region in the other color analysis method is the ranking position of each comparison region in the sorting of the color comparison region corresponding to the other color analysis method.

Therefore, according to the technical content of the case, the case can provide the user to select the comparison area closest to the color of the sample area from the different comparison area rankings provided after comparing the sample areas by different color analysis methods.

100‧‧‧Color comparison processing platform

102‧‧‧Image capture device

104‧‧‧ processor

106‧‧‧ Database

108‧‧‧Test paper

1082‧‧‧Sample area

1061-1065‧‧‧ Comparison area

300‧‧‧Flow

301-307‧‧‧Step

Figure 1 shows a color comparison processing system for detecting test paper according to an embodiment of the present invention.

Figure 2 shows the sorting results after using three different color analysis systems to confirm the color of the test strip sample area and the color of the comparison area after comparison.

FIG. 3 is a flowchart of selecting a color comparison area according to an embodiment of the present invention.

The following will clearly illustrate the spirit of the case with diagrams and detailed descriptions. Anyone with ordinary knowledge in the technical field who understands the embodiments of the case will be able to change and modify the techniques taught in the case without departing from the case. Spirit and scope.

The terminology in this article is only for describing specific embodiments, and is not intended to be a limitation of this case. Singular forms such as "a", "this", "this", "this" and "the", as used herein, also include plural forms.

As used herein, "coupled" or "connected" can refer to two or more elements or devices making direct physical contact with each other, or indirectly making physical contact with each other, or can refer to two or more elements or devices making mutual contact Operation or action.

The terms "contains", "includes", "has", "contains", etc. used in this article are all open terms, which means including but not limited to.

As used herein, "and/or" includes any or all combinations of the things described.

Regarding the terms used in this article, unless otherwise noted, they usually have the ordinary meaning that each term is used in this field, in the content of this case, and in special content. Certain terms used to describe this case will be discussed below or elsewhere in this specification to provide additional guidance to those skilled in the art in the description of this case.

Figure 1 shows a color comparison processing system for detecting test paper according to an embodiment of the present invention. The color comparison processing platform 100 includes an image capturing device 102, a processor 104, and a database 106. The image capturing device 102 is used to photograph a test paper 108 and generate an image data. Test strip 108 includes Same as this area 1082. The sample area 1082 is used to receive the subject's body fluid, such as urine, for testing. The sample area 1082 is made of a color-changing reaction material. When the sample area 1082 receives body fluids, the color changes. After the image capture device 102 uploads the image data of the test strip 108, the processor 104 judges the index range of each test item by comparing the color of the sample area 1082 and the comparison area 1061-1065 in the database 106 for the purpose of discovering in advance Whether there are related diseases, so as to discover the health status of the subject in time and seek medical treatment as soon as possible. It is worth noting that in this embodiment, only five comparison regions 1061-1065 are used for comparison with the sample region 1082 of the test strip 108, but the number of comparison regions is not limited to this embodiment. However, when the color comparison processing platform 100 performs the color comparison between the sample area 1082 and the comparison area 1061-1065, the comparison result will be different due to the different color analysis methods used, and there will be different interpretation results, resulting in The user is at a loss. Therefore, the present invention provides a scoring mechanism for scoring the analysis results of different color analysis methods, so that users can evaluate the analysis results of different color analysis methods based on the scoring, and then select the closest sample area from them To determine the index range of the test item.

The figure 2 assumes that the processor uses three different color analysis systems to confirm the color of the test strip sample area and the color of the comparison area. In one embodiment, the color analysis method used may use RGB color space, CIEDE2000 color space or HSV color space for color analysis. However, in other embodiments, other color analysis methods, such as the Delta E, CIE76 or CIE94 color analysis system, can also be used. Among them, the first color analysis method uses RGB color space for color analysis, the second color analysis method uses CIEDE2000 color space for color analysis, and the third color analysis method uses HSV color space for color analysis. According to this, as shown in Figure 2, according to the first color analysis method, the sample area 1082 and the comparison area 1061-1065 are color-matched, and the color similarity is ranked from the most similar to the most dissimilar. The order is the comparison region 1064, the comparison region 1063, the comparison region 1062, the comparison region 1065, and the comparison region 1061. According to the second color analysis method, the color comparison between the sample area 1082 and the comparison area 1061-1065, the color similarity, from the most similar to the most dissimilar, the ranking result is the comparison area 1063, the comparison area 1064. Comparison region 1062, comparison region 1061, and comparison region 1065. According to the third color analysis method, the color comparison between the sample area 1082 and the comparison area 1061-1065 is performed. The color similarity ranges from the most similar to the most dissimilar. The ranking results are the comparison area 1064 and the comparison area. 1062, comparison region 1063, comparison region 1061, and comparison region 1065. In this embodiment, the ranking result of the comparison region is the score of the comparison region. For example, if a comparison region is ranked as 5, then the comparison region's score is 5; if the comparison region is ranked as 1, then this The score of the comparison area is 1, and so on.

After obtaining these sorting results, each color analysis method can be scored. In one embodiment, the present invention uses the discounted cumulative gain method (Discounted Cumulative Gain, DCG) to calculate each ratio in each color analysis system. The total score for the area. Through this method, the differences of different scoring systems in different comparison regions can be considered. It is further known that the scoring results of those color analysis systems are relatively close. The formula used is as follows:

Among them, K represents the k-th comparison region; Color_system_score _{i, k} represents the score of the _i- th color analysis method in the k-th color comparison region.

Accordingly, as shown in Figure 2, the first color analysis method is to use RGB color space for color analysis. Under this analysis method, the comparison area 1061 is the first comparison area, so its K=1, and this The ranking of the first comparison area is 5, so the score is 5, which is Color_system_score _i,k =5. The comparison area 1062 is the second comparison area, so its K=2, and the ranking of this second comparison area is 3, so the score is 3, that is, Color_system_score _i,k =3. The comparison region 1063 is the third comparison region, so its K=3, and the ranking of the third comparison region is 2, so the score is 2, which is Color_system_score _i,k =2. The comparison area 1064 is the fourth comparison area, so its K=4, and the ranking of the fourth comparison area is 1, so the score is 1, which is Color_system_score _i,k =1. The comparison area 1065 is the fifth comparison area, so its K=5, and the ranking of the fifth comparison area is 4, so the score is 4, which is Color_system_score _i,k =4. Accordingly, the total score of the discounted cumulative gain method of the first color analysis method is:

The second color analysis method is to use CIEDE2000 color space for color analysis. Under this analysis method, the comparison area 1061 is the first comparison area, so its K=1, and the ranking of the first comparison area is 4, so the score is 4, which is Color_system_score _i,k =4. The comparison area 1062 is the second comparison area, so its K=2, and the ranking of this second comparison area is 3, so the score is 3, that is, Color_system_score _i,k =3. The comparison area 1063 is the third comparison area, so its K=3, and the ranking of the third comparison area is 1, so the score is 1, which is Color_system_score _i,k =1. The comparison area 1064 is the fourth comparison area, so its K=4, and the ranking of the fourth comparison area is 2, so the score is 2, which is Color_system_score _i,k =2. The comparison area 1065 is the fifth comparison area, so its K=5, and the ranking of the fifth comparison area is 5, so the score is 5, which is Color_system_score _i,k =5. Accordingly, the total score of the discounted cumulative gain method of the second color analysis method is:

The third color analysis method is to use HSV color space for color analysis. Under this analysis method, the comparison area 1061 is the first comparison area, so its K=1, and the ranking of the first comparison area is 4, so the score is 4, which is Color_system_score _i,k =4. The comparison region 1062 is the second comparison region, so its K=2, and the ranking of this second comparison region is 2, so the score is 2, which is Color_system_score _i,k =2. The comparison area 1063 is the third comparison area, so its K=3, and the ranking of the third comparison area is 3, so the score is 3, that is, Color_system_score _i,k =3. The comparison area 1064 is the fourth comparison area, so its K=4, and the ranking of the fourth comparison area is 1, so the score is 1, which is Color_system_score _i,k =1. The comparison area 1065 is the fifth comparison area, so its K=5, and the ranking of the fifth comparison area is 5, so the score is 5, which is Color_system_score _i,k =5. Accordingly, the total score of the discounted cumulative gain method of the third color analysis method is:

Accordingly, the scores of the second color analysis method and the third color analysis method are relatively close. In other words, the color analysis results obtained by using the CIEDE2000 color space for sample area color analysis and the HSV color space for sample area color analysis are relatively close, because the same sample area can still be obtained by two different color analysis methods. Analysis results, so users can confirm that the analysis results are more correct.

In order to further evaluate which sample area is closest to the color of the comparison area, the three color analysis methods will be given unequal weights according to the above scoring results, because the second color analysis method and the third color analysis The scores of the methods are relatively close. In order to highlight the importance of these two color analysis methods, the second color analysis method and the third color analysis method are given a higher weight, and the inconsistent color analysis methods, that is, The first color analysis method gives relatively low weight. In an embodiment, the weight given to the second color analysis method and the third color analysis method is 0.4, and the weight given to the first color analysis method is 0.2. However, in other embodiments, different weight values may be given. Accordingly, the color scoring formula of the comparison area is: the color score of the comparison area = 0.2 * (the comparison area score in the first color analysis method) + 0.4 * (the comparison in the second color analysis method Area score) +0.4 * (This comparison area score in the third color analysis method)

As shown in Figure 2, for the area 1061, the first color analysis method score is 5, the second color analysis method score is 4, and the third color analysis method score is 4. Therefore, the comparison area The color score of 1061 = 0.2*5+0.4*4+0.4*4=4.2. Compared with the area 1062, the first color analysis method score is 3, the second color analysis method score is 3, and the third color analysis method score is 2. Therefore, the color score of the comparison area 1062 = 0.2 * 3+0.4 *3+0.4 * 2=2.6. Compared to the area 1063, its One color analysis method has a score of 2, the second color analysis method has a score of 1, and the third color analysis method has a score of 3. Therefore, the color score of the comparison area 1063 = 0.2 * 2+0.4 * 1 + 0.4 * 3 = 2 . For the area 1064, the first color analysis method score is 1, the second color analysis method score is 2, and the third color analysis method score is 1. Therefore, the color score of the comparison area 1064 = 0.2 * 1+0.4 *2+0.4 * 1=1.4. For the area 1065, the first color analysis method score is 4, the second color analysis method score is 5, and the third color analysis method score is 5. Therefore, the color score of the comparison area 1065 = 0.2 * 4+0.4 *5+0.4 *5=4.8. Among them, the score of the comparison area 1064 is 1.4, which is the lowest, and the color of the comparison area 1064 is closest to the color of the sample area.

FIG. 3 is a flowchart of selecting a color comparison area according to an embodiment of the present invention. Process 300, first in step 301, according to the same region, use a plurality of color analysis methods to analyze the color difference between the plurality of color comparison regions and the sample region and sort them to form a plurality of color comparison regions. A color analysis method, in which the sample area is used to receive a subject's body fluid and correspondingly display a color. In an embodiment, as shown in FIGS. 1 and 2, the first color analysis method, the second color analysis method, and the third color analysis method are used to test the sample area 1082 and the comparison area 1061-1065, respectively. The color comparison is performed to form a plurality of color comparison regions, which respectively correspond to the first color analysis method, the second color analysis method, and the third color analysis method. Next, in step 302, the scores of each of the plurality of color analysis methods are calculated. In one embodiment, the discounted cumulative gain method (Discounted Cumulative Gain, DCG) is used to calculate the scores of each color analysis method. The formula used is as follows:

Among them, K represents the kth comparison area; Color_system_score _{i, k} represents the score of the _ith color analysis method in the kth color comparison area.

When the score of each color analysis method is calculated, in step 303, at least two color analysis methods with the closest scores among the color analysis methods are selected, and in step 304, the at least two color analysis methods are set to have a first Weights, and set the remaining color analysis methods to have other weights different from the first weight, where the first weight is greater than the other weights. In an embodiment, if the second color analysis method among the three color analysis methods is close to the discounted cumulative gain method score of the third color analysis method, in this case, the second color analysis method and the third This color analysis method has a relatively high first weight, and the first color analysis method, which is relatively inconsistent, sets a weight other than this first weight.

Then in step 305, the score of each comparison area is calculated according to the first weight and other weights. In an embodiment, if the set first weight is 0.4, the other weights are 0.2. At this time, the color score of the comparison area = 0.2 * (the score of the comparison area in the first color analysis method) + 0.4 * (the score of the comparison area in the second color analysis method) + 0.4 * (the third In the color analysis method, this comparison area score). Next, in step 306, the score of each comparison area is compared, so that in step 307, the lowest of the scores of each comparison area is selected as the comparison area that is closest to the color presented by the sample area.

Based on the above, this case can provide users to select the comparison area closest to the color of the sample area from the different comparison area rankings provided after comparing the sample areas in different color analysis methods.

Although the case is disclosed as above with examples, it is not used to limit the case. Anyone who is familiar with this skill can make various changes and retouching without departing from the spirit and scope of the case, so the scope of protection of the case should be attached The scope defined in the scope of patent application shall prevail.

300‧‧‧Flow

301-307‧‧‧Step

Claims (6)

A color comparison area selection method is used to select a comparison area from the plurality of color comparison areas that is closest to the color presented by the same area. The method at least includes: according to the same area, use a plurality of color analysis The method analyzes the color difference between the plurality of color comparison areas and the sample area and sorts them to form a plurality of color comparison areas. The sequence corresponds to the plurality of color analysis methods, wherein the sample area is used to receive a subject's body fluid and Corresponding to displaying a color; calculating the score of each of the plurality of color analysis methods; selecting at least two color analysis methods with the closest scores among the color analysis methods; setting the at least two color analysis methods to have a first weight, And setting the remaining color analysis methods to have other weights different from the first weight, wherein the first weight is greater than the other weights; calculate each comparison area score according to the first weight and the other weights; compare each The score of the comparison area; and selecting the lowest of the scores of each comparison area to be a comparison area that is closest to the color presented by the sample area, wherein the score of each of the plurality of color analysis methods is calculated using the cumulative discount Calculated by the method of Discounted Cumulative Gain (DCG), where:

Among them, K represents the k-th color comparison region; Color_system_score _{i, k} represents the score of the _i- th color analysis method in the k-th color comparison region. The color comparison area selection method according to claim 1, wherein the sample area is used to receive urine from a subject. The method for selecting a color comparison region according to claim 1, wherein the score of the i-th color analysis method in the k-th color comparison region is that the k-th color comparison region is analyzed in the i-th color The sorting position in the sorting of the color comparison area corresponding to the method. The method for selecting a color comparison region according to claim 1, wherein calculating the score of each comparison region according to the first weight and the other weights further comprises: multiplying the first weight by the at least two color analysis methods respectively Each of the comparison area scores obtains a first weight type score; the other weights are respectively multiplied by the corresponding each of the comparison area scores in the other color analysis methods to obtain other weight type scores; and the total A weight type score and the other weight type scores are used as the scores of each comparison area. The method for selecting a color comparison area according to claim 4, wherein the score of each comparison area in the at least two color analysis methods is the corresponding of each comparison area in the at least two color analysis methods Sorting position in the color comparison area sorting. The method for selecting a color comparison area according to claim 4, wherein the score of each comparison area in the other color analysis method is the ratio of each The sorting position of the region in the color comparison region corresponding to the other color analysis methods.

Images