Method and device for identifying browning resistance of pulp

Abstract

The invention discloses a pulp browning resistance identification method and device, and relates to the technical field of fruit detection.The method comprises the steps that pulp sections of different varieties are obtained, and original images of the pulp sections are collected; respectively collecting pulp section images of each pulp section at different time according to a time sequence; calculating the Lab value of each pulp section image, and calculating the browning index of each pulp section image according to the Lab value of each pulp section image; the Lab value comprises brightness, red-green and yellow-blue components; segmenting the browning area of each pulp section image, extracting the browning area of each pulp section image, extracting browning segmentation pixels from the browning areas, and determining the browning area of each pulp section image; and according to the Lab value change value, the browning index and the browning area of the pulp section image at different time, calculating and determining the browning resistance ranking of the pulp of different varieties by using a browning membership function. According to the method, the browning identification speed and accuracy of the pulp can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of fruit detection, and in particular to a method and device for identifying fruit flesh browning resistance. Background Art

[0002] At present, with the accelerated pace of life, fresh-cut fruits are increasingly accepted by consumers for their health, convenience, and 100% edible characteristics. The development of fresh-cut fruit products is directly related to the browning resistance of the flesh tissue. For example, loquat fresh fruit needs to be peeled and eaten immediately. After cutting, the flesh is less resistant to browning, which does not meet the modern fast-paced consumer needs. Identifying and exploring varieties with browning resistance and cultivating varieties with browning resistance are new highlights for increasing the proportion of consumers of loquat fresh fruit products and promoting the sustainable development of the loquat industry. At present, the evaluation and analysis of fruit color traits often uses a colorimeter for detection. Among them, a method for determining the browning degree of the flesh has published a method for identifying the browning resistance of apple, peach or pear flesh using a colorimeter, but the improper fitting of the caliber of the colorimeter is prone to deviation, and multi-point detection is not conducive to rapid batch detection and reducing sampling errors. Summary of the invention

[0003] The purpose of the present application is to provide a method and device for identifying fruit pulp browning resistance, which can improve the speed and accuracy of fruit pulp browning identification.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a method for identifying resistance to browning of fruit pulp, comprising:

[0006] Obtain the flesh sections of different varieties and collect the original images of each flesh section;

[0007] Collect the images of the fruit flesh sections at different times in chronological order;

[0008] Calculating the Lab value of each of the flesh section images, and calculating the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component, and yellow-blue component;

[0009] Extracting the browning region of each of the flesh section images, and extracting browning segmentation pixels from each of the browning regions, and determining the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels;

[0010] According to the Lab value change, browning index and browning area of ​​the flesh cross-section images at different times, the browning membership function was calculated to determine the browning resistance ranking of the flesh of different varieties.

[0011] Optionally, calculating the Lab value of each of the pulp cross-section images specifically includes:

[0012] MATLAB is used to read the RGB channels of the pulp cross-section image, the RGB channels are converted into a CIE-Lab space, and the Lab value is extracted from the CIE-Lab space.

[0013] Optionally, calculating the browning index of each of the pulp section images according to the Lab value of each of the pulp section images specifically includes:

[0014] According to the formula Calculating the browning index of each of the pulp section images;

[0015] Among them, BI is the browning index, L n is the brightness of the pulp cross-section image at time n, L0 is the brightness of the original image, a n is the red and green component of the flesh cross-section image at time n, a0 is the red and green component of the original image, b n is the yellow-blue component of the pulp cross-section image at time n, and b0 is the yellow-blue component of the original image.

[0016] Optionally, extracting the browning region of each of the flesh section images, extracting browning segmentation pixels from each of the browning regions, and determining the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels specifically includes:

[0017] For one of the pulp cross-section images, the browning region in the pulp cross-section image is extracted using the ROI function in MATLAB;

[0018] Calculating the pixel average value of the browning area;

[0019] sequentially calculating the Euclidean distance between each pixel in the browning area and the pixel average value, and dividing the pixels whose Euclidean distance is less than or equal to a set threshold into browning segmentation pixels;

[0020] The browning area of ​​the flesh section image is determined according to each browning segmentation pixel.

[0021] Optionally, according to the Lab value change value, browning index and browning area of ​​the flesh section image at different times, the browning membership function is calculated and the browning resistance ranking of the flesh of different varieties is determined, specifically including:

[0022] The principal component analysis method is used to extract the principal component index from multiple component indexes; the multiple component indexes include the Lab value change value, browning index and browning area of ​​the flesh section image at different times;

[0023] The membership function value of each principal component index is calculated using the membership function formula;

[0024] The membership function values ​​of the principal component indicators corresponding to each variety of fruit are weighted and summed to obtain the comprehensive score of each variety of fruit;

[0025] Based on the comprehensive score of the fruits of each variety, the ranking of browning resistance of the flesh of each variety was determined.

[0026] Optionally, the membership function formula is expressed as: μ(X ij )=(X ij -X min ) / (X max -X min );

[0027] Among them, μ(X ij ) is the membership function value of the jth principal component index of the ith variety, X ij is the jth principal component index of the ith variety, X max is the maximum value of the jth principal component index, X min is the minimum value of the jth principal component index.

[0028] Optionally, the fruit is loquat.

[0029] Optionally, before obtaining the flesh sections of different varieties and collecting the original images of each flesh section, the method for identifying flesh browning resistance further comprises:

[0030] Arrange fruits of different varieties in a matrix in a photographic light box; the number of fruit rows in the matrix is ​​≤4, the number of columns is ≤8, the fruits are arranged at intervals of 2 to 5 cm, the fruit flesh sections are kept horizontal and the heights of the fruit flesh sections are consistent, the width of the matrix is ​​≤30 cm and the length is ≤50 cm.

[0031] Optionally, collecting the flesh section images of each flesh section at different times in chronological order specifically includes:

[0032] If the loquat is white flesh, collect the flesh section images of each flesh section at 10min, 30min and 60min in chronological order; the time for collecting the original image of each flesh section is 0min;

[0033] If the loquat is a red flesh type, the flesh section images of each flesh section are collected at 30 minutes and 60 minutes in chronological order; the time for collecting the original image of each flesh section is 0 minutes.

[0034] In a second aspect, the present application provides a device for identifying resistance to browning of fruit pulp, comprising:

[0035] The original image acquisition module is used to obtain the flesh sections of different varieties and collect the original images of each flesh section;

[0036] The flesh section image acquisition module is used to respectively acquire the flesh section images of each flesh section at different times in chronological order;

[0037] A browning index calculation module, used to calculate the Lab value of each of the flesh section images, and calculate the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component and yellow-blue component;

[0038] A browning area calculation module is used to extract the browning area of ​​each of the flesh section images, extract browning segmentation pixels from each of the browning areas, and determine the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels;

[0039] The browning ranking module is used to calculate the browning membership function according to the Lab value change, browning index and browning area of ​​the flesh cross-section image at different times, and determine the browning resistance ranking of the flesh of different varieties.

[0040] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0041] The present application provides a method and device for identifying browning resistance of fruit flesh, which extracts the browning region of each of the fruit flesh cross-section images, and extracts browning segmentation pixels from each of the browning regions, and determines the browning area of ​​each of the fruit flesh cross-section images based on the extracted browning segmentation pixels, thereby achieving pixel-level image segmentation, which can improve the accuracy of browning identification of the fruit flesh. In addition, the browning identification is performed without relying on a colorimeter, and the browning identification is performed by acquiring images and processing images, thereby enabling batch detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A schematic diagram of a process for identifying fruit flesh browning resistance provided in one embodiment of the present application;

[0044] Figure 2 A schematic diagram of image processing in the process of calculating the Lab value of a pulp cross-section image provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of segmentation effects of different segmentation thresholds provided in an embodiment of the present application;

[0046] Figure 4A schematic diagram of the functional modules of a device for identifying resistance to fruit browning provided in one embodiment of the present application. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0048] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0049] The present application provides a method for identifying resistance to flesh browning. Figure 1 As shown, the method for identifying the resistance to browning of fruit flesh comprises:

[0050] Step 101: Obtain the flesh cross sections of different varieties and collect the original images of each flesh cross section.

[0051] Step 102: Collect the images of the fruit flesh sections at different times in chronological order.

[0052] Step 103: Calculate the Lab value of each of the flesh section images, and calculate the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component, and yellow-blue component.

[0053] Step 104: extracting the browning region of each of the flesh section images, extracting browning segmentation pixels from each of the browning regions, and determining the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels.

[0054] Step 105: Calculate the browning membership function according to the Lab value change value, browning index and browning area of ​​the flesh cross-section image at different times, and determine the browning resistance ranking of the flesh of different varieties.

[0055] The browning region of each of the flesh cross-section images is extracted, and browning segmentation pixels are extracted from each browning region. The browning area of ​​each of the flesh cross-section images is determined according to the extracted browning segmentation pixels, thereby realizing pixel-level image segmentation. This can improve the accuracy of browning identification of the flesh. In addition, browning identification is performed without relying on a colorimeter, and browning identification is performed by acquiring images and processing images, thereby enabling batch detection.

[0056] In an exemplary embodiment, the fruit is loquat. The loquat flesh browning resistance identification method of the present application is used to identify the loquat flesh browning resistance, which specifically includes: loquat fruit picking, cutting, photographing, data collection, and data analysis. The specific operation is as follows: in the loquat fruit ripening season, for each loquat germplasm resource, fruits with consistent crown outer size, color and maturity, no pests and diseases, and no mechanical damage are picked and brought back to the laboratory in time. In the laboratory, the fruits are cut into two equal parts longitudinally, and the second half after removing the seeds and carpels is placed in a soft light photography light box with a fixed light source. A camera is used to record the browning phenotype at different times after fresh cutting, that is, the original images of each flesh section and the flesh section images at different times are collected; the RGB channels of the image are read using MATLAB software, and converted to CIE-Lab space, the Lab value is extracted, and the browning index is calculated; the color feature values ​​of the representative areas of loquat flesh browning are extracted using MATLAB software, and the browning area segmentation of the flesh section image is completed using Euclidean distance; the browning membership function is calculated to determine the browning grade of different varieties of loquat. The other half of the fruit was used for the colorimeter method (D65 light source 10° field of view) to measure the Lab value and RGB value of the 0min and 60min sections for comparison with the method of this application. The temperature of the observation environment was maintained at 25°C.

[0057] Before step 101, the method for identifying resistance to fruit flesh browning further includes: placing different varieties of fruits in a matrix in a photographic light box; the number of fruit rows in the matrix is ​​≤4, the number of columns is ≤8, the fruits are placed at an interval of 2 to 5 cm, the fruit flesh sections are kept horizontal and the heights of the fruit flesh sections are consistent, the placement directions are consistent, the width of the matrix is ​​≤30 cm and the length is ≤50 cm; the length and width of the tray for placing the fruits are 50 cm to 70 cm, and are covered with a gray background cloth.

[0058] The length, width and height of the photography light box range from 50cm to 70cm. The photography light box is supplemented with light by a non-stroboscopic white light LED light board. The observation environment temperature of the photography light box is 25℃.

[0059] Among them, step 102 specifically includes: if the loquat is a white flesh type, the flesh section images of each flesh section are collected at 10min, 30min and 60min in chronological order; the time for collecting the original image of each flesh section is 0min, that is, the original image is collected 0min after the fruit is freshly cut; if the loquat is a red flesh type, the flesh section images of each flesh section are collected at 30min and 60min in chronological order, and the time for collecting the original image of each flesh section is 0min.

[0060] The distance between the camera for collecting the original image and the pulp cross-section images at different times and the fruit cross-section is 50 cm to 70 cm; the camera parameters are aperture value F=5.6, sensitivity ISO=1000, shutter speed 640, and image accuracy 1600w pixels (inclusive) or above.

[0061] The images of the fruit pulp sections at different times are obtained by purifying the gray background into a black background using Photoshop software.

[0062] The step 103 of calculating the Lab value of each of the pulp cross-section images specifically includes: using MATLAB to read the RGB channels of the pulp cross-section images, converting the RGB channels into a CIE-Lab space, and extracting the Lab value from the CIE-Lab space.

[0063] Using MATLAB to read the RGB channels of the pulp cross-section image, converting the RGB channels into CIE-Lab space, and extracting Lab values ​​from the CIE-Lab space, specifically including:

[0064] The imread function is used to read the RGB channels of the pulp cross-section image.

[0065] Use the rgb2lab function to convert the RGB channels to CIE-Lab space.

[0066] The CIE-Lab space ( Figure 2 (a)) is converted to grayscale ( Figure 2 (b) of the report.

[0067] The segmentation threshold of the grayscale image is calculated using the Sobel operator.

[0068] Using the edge function, a binary mask image of the pulp section image is obtained according to the segmentation threshold ( Figure 2 Part (c) of the figure). The Sobel operator segmentation threshold is 0.04 to 0.06.

[0069] The binary mask shows high-contrast lines in the image, but these lines do not depict the outline of the fruit slice object well. Therefore, the imdilate function is used to perform morphological dilation on the binary mask image to obtain the outline image of the fruit slice ( Figure 2 (d) in the

[0070] The imfill function is used to fill the pores of the contour image of the pulp slice to obtain the filled image ( Figure 2 (e) of the annex).

[0071] Use the imclearborder function to delete the border connected objects and get the image after deleting the border connected objects ( Figure 2 (f) of the above paragraphs.

[0072] The imerode function is used for smoothing to obtain a complete binary segmentation image with a background of 0 and a target area of ​​1 ( Figure 2 (g) in the above table.

[0073] The white area with a value of 1 is mapped to the Lab channel in the original image, and the Lab mean of the pulp section is calculated, that is, the extracted Lab value is obtained.

[0074] The rgb2gray function, Sobel operator, edge function, imdilate function, imfill function, imclearborder function, and imerode function are all functions in the MATLAB software.

[0075] In step 103, the browning index of each of the pulp section images is calculated according to the Lab value of each of the pulp section images, specifically comprising: according to the formula Calculate the browning index of each of the pulp cross-section images.

[0076] Among them, BI is the browning index, L n is the brightness of the pulp cross-section image at time n, L0 is the brightness of the original image, a n is the red and green component of the flesh cross-section image at time n, a0 is the red and green component of the original image, b n is the yellow-blue component of the pulp cross-section image at time n, and b0 is the yellow-blue component of the original image.

[0077] Among them, step 104 uses the ROI function in MATLAB to select an irregular representative loquat flesh browning area, extracts color feature values, and uses the Euclidean distance to complete the browning area segmentation of the original image, specifically including: for a flesh section image, using the ROI function in MATLAB to extract the browning area in the flesh section image; calculating the standard deviation of the red component, the standard deviation of the green component, and the standard deviation of the blue component of the browning area; determining the gray value T according to the maximum value of the standard deviation of the red component, the standard deviation of the green component, and the standard deviation of the blue component, specifically: taking the approximate value of the three standard deviations as the gray value T; calculating the pixel average value of the browning area; sequentially calculating the Euclidean distance between each pixel in the browning area and the pixel average value, and dividing the pixels whose Euclidean distance is less than or equal to the set threshold as browning segmentation pixels; determining the browning area of ​​the flesh section image according to each browning segmentation pixel. The threshold is set to be several times the gray value T.

[0078] The T value is a fixed grayscale value. During segmentation, the pixels whose grayscale values ​​are less than or equal to the threshold T are set to 1 and represented by white, and the pixels whose grayscale values ​​are greater than the threshold T are set to 0 and represented by black.

[0079] Figure 3Part (a) is a schematic diagram of the ROI area of ​​white flesh. Figure 3 Part (b) is a schematic diagram of the cross section of the Guifei variety of loquat flesh at 60 minutes. Figure 3 Part (i) in the middle is a schematic diagram of the cross-section of the golden nugget variety of loquat pulp at 60 minutes. Figure 3 Parts (c) to (g) are Figure 3 Part (b) shows the segmentation effect when 70T, 80T, 90T, 100T, and 110T are the segmentation thresholds. Figure 3 Part (j) - Part (n) are Figure 3 Part (i) shows the segmentation results when 10T, 15T, 20T, 25T, and 30T are the segmentation thresholds.

[0080] The browning of loquat flesh is uneven. When selecting the target ROI area, try to choose an area containing different gradients of browning. The color value of white flesh is obviously different from that of browning, so the T value used for segmentation is relatively large. 90T is a relatively suitable segmentation threshold, such as Figure 3 Part (c) to part (g); the browning of red flesh is manifested as a deepening of flesh color, and the T value used for segmentation is small, such as Figure 3 In the middle (j)-(n) part, when selecting the ROI area of ​​the red meat type, try to choose the area with the most significant color difference from the flesh. T is the set gray value.

[0081] The browning area segmentation threshold of the ROI function is 80T-100T for white meat varieties and 15T-25T for red meat varieties.

[0082] Wherein, step 105 specifically includes: extracting the principal component index from the plurality of component indexes by using the principal component analysis method; the plurality of component indexes include the Lab value change value, the browning index and the browning area of ​​the pulp cross-section image at different times.

[0083] The membership function formula is used to calculate the membership function value of each principal component index.

[0084] The membership function values ​​of the principal component indicators corresponding to each variety of fruit are weighted and summed to obtain the comprehensive score of each variety of fruit.

[0085] Based on the comprehensive score of the fruits of each variety, the ranking of browning resistance of the flesh of each variety was determined.

[0086] The membership function formula is expressed as: μ(X ij )=(X ij -X min ) / (X max -X min ).

[0087] Among them, μ(X ij ) is the membership function value of the jth principal component index of the ith variety, X ij is the jth principal component index of the ith variety, X max is the maximum value of the jth principal component index, X min is the minimum value of the jth principal component index.

[0088] According to the changes in Lab values ​​during the browning process, the browning index (BI) and browning area (S) at different times, the browning membership function rankings of the 10 germplasm resources were calculated using principal component analysis and membership function analysis, as shown in Table 1.

[0089] Table 1 Ranking of browning membership functions of 110 germplasm resources

[0090]

[0091]

[0092] This application can take browning phenotype images of fresh-cut fruits in batches, quickly and efficiently, and obtain color phenotype data of the cut surface through post-analysis, which significantly improves the efficiency and accuracy of identification. Compared with the conventional colorimeter measurement method and naked eye sensory evaluation, the method of this application can accurately identify browning of the flesh to the pixel level, and can realize batch, fast and efficient identification of loquat germplasm resources for flesh resistance, which is an important technical support for the excavation and innovative use of browning-resistant germplasm (variety).

[0093] Based on the same inventive concept, the embodiment of the present application also provides a flesh browning resistance identification device for implementing the above-mentioned flesh browning resistance identification method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations of one or more flesh browning resistance identification device embodiments provided below can refer to the limitations of the flesh browning resistance identification method above, and will not be repeated here.

[0094] In an exemplary embodiment, Figure 4 As shown, a device for identifying resistance to browning of fruit pulp is provided, comprising:

[0095] The original image acquisition module is used to obtain the flesh sections of different varieties and collect the original images of each flesh section.

[0096] The flesh section image acquisition module is used to respectively acquire the flesh section images of each flesh section at different times in chronological order.

[0097] The browning index calculation module is used to calculate the Lab value of each of the flesh section images, and calculate the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component and yellow-blue component.

[0098] The browning area calculation module is used to extract the browning area of ​​each of the flesh section images, extract the browning segmentation pixels from each of the browning areas, and determine the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels.

[0099] The browning ranking module is used to calculate the browning membership function according to the Lab value change, browning index and browning area of ​​the flesh cross-section image at different times, and determine the browning resistance ranking of the flesh of different varieties.

[0100] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for identifying resistance to flesh browning, characterized in that: The method for identifying flesh browning resistance comprises: Obtain the flesh sections of different varieties and collect the original images of each flesh section; Collect the images of the fruit flesh sections at different times in chronological order; Calculating the Lab value of each of the flesh section images, and calculating the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component, and yellow-blue component; Extracting the browning region of each of the flesh section images, and extracting browning segmentation pixels from each of the browning regions, and determining the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels; According to the Lab value change, browning index and browning area of ​​the flesh cross-section images at different times, the browning membership function was calculated to determine the browning resistance ranking of the flesh of different varieties.

2. The method for identifying flesh browning resistance according to claim 1, characterized in that: Calculating the Lab value of each of the pulp cross-section images specifically includes: MATLAB is used to read the RGB channels of the pulp cross-section image, the RGB channels are converted into a CIE-Lab space, and Lab values ​​are extracted from the CIE-Lab space.

3. The method for identifying flesh browning resistance according to claim 1, characterized in that: Calculating the browning index of each of the pulp section images according to the Lab value of each of the pulp section images specifically includes: According to the formula Calculating the browning index of each of the pulp section images; Among them, BI is the browning index, L n is the brightness of the pulp cross-section image at time n, L0 is the brightness of the original image, a n is the red and green component of the flesh cross-section image at time n, a0 is the red and green component of the original image, b n is the yellow-blue component of the pulp cross-section image at time n, and b0 is the yellow-blue component of the original image.

4. The method for identifying flesh browning resistance according to claim 1, characterized in that: Extracting the browning region of each of the flesh section images, extracting browning segmentation pixels from each of the browning regions, and determining the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels, specifically includes: For one of the pulp cross-section images, the browning region in the pulp cross-section image is extracted using the ROI function in MATLAB; Calculating the pixel average value of the browning area; sequentially calculating the Euclidean distance between each pixel in the browning area and the pixel average value, and dividing the pixels whose Euclidean distance is less than or equal to a set threshold into browning segmentation pixels; The browning area of ​​the flesh section image is determined according to each browning segmentation pixel.

5. The method for identifying resistance to flesh browning according to claim 1, characterized in that: According to the Lab value change, browning index and browning area of ​​the flesh section images at different times, the browning membership function is calculated to determine the browning resistance ranking of the flesh of different varieties, including: The principal component analysis method is used to extract the principal component index from multiple component indexes; the multiple component indexes include the Lab value change value, browning index and browning area of ​​the flesh section image at different times; The membership function value of each principal component index is calculated using the membership function formula; The membership function values ​​of the principal component indicators corresponding to each variety of fruit are weighted and summed to obtain the comprehensive score of each variety of fruit; Based on the comprehensive score of the fruits of each variety, the ranking of browning resistance of the flesh of each variety was determined.

6. The method for identifying resistance to flesh browning according to claim 1, characterized in that: The membership function formula is expressed as: μ(X ij )=(X ij -X min ) / (X max -X min ); Among them, μ(X ij ) is the membership function value of the jth principal component index of the ith variety, X ij is the jth principal component index of the ith variety, X max is the maximum value of the jth principal component index, X min is the minimum value of the jth principal component index.

7. The method for identifying resistance to flesh browning according to claim 1, characterized in that: The fruit is loquat.

8. The method for identifying resistance to flesh browning according to claim 7, characterized in that: Before obtaining the flesh sections of different varieties and collecting the original images of each flesh section, the method for identifying flesh browning resistance further includes: Arrange fruits of different varieties in a matrix in a photographic light box; the number of fruit rows in the matrix is ​​≤4, the number of columns is ≤8, the fruits are arranged at intervals of 2 to 5 cm, the fruit flesh sections are kept horizontal and the heights of the fruit flesh sections are consistent, the width of the matrix is ​​≤30 cm and the length is ≤50 cm.

9. The method for identifying resistance to flesh browning according to claim 7, characterized in that: The images of the flesh sections of each flesh section at different times are collected in chronological order, specifically including: If the loquat is white flesh, collect the flesh section images of each flesh section at 10min, 30min and 60min in chronological order; the time for collecting the original image of each flesh section is 0min; If the loquat is a red flesh type, the flesh section images of each flesh section are collected at 30 minutes and 60 minutes in chronological order; the time for collecting the original image of each flesh section is 0 minutes.

10. A device for identifying resistance to flesh browning, characterized in that: The flesh browning resistance identification device comprises: The original image acquisition module is used to obtain the flesh sections of different varieties and collect the original images of each flesh section; The flesh section image acquisition module is used to respectively acquire the flesh section images of each flesh section at different times in chronological order; A browning index calculation module, used to calculate the Lab value of each of the flesh section images, and calculate the browning index of each of the flesh section images according to the Lab value of each of the flesh section images; the Lab value includes brightness, red-green component and yellow-blue component; A browning area calculation module is used to extract the browning area of ​​each of the flesh section images, extract browning segmentation pixels from each of the browning areas, and determine the browning area of ​​each of the flesh section images according to the extracted browning segmentation pixels; The browning ranking module is used to calculate the browning membership function according to the Lab value change, browning index and browning area of ​​the flesh cross-section image at different times, and determine the browning resistance ranking of the flesh of different varieties.

Images