A rapid scanning detection method for the distribution of biogenic amine content in fish meat

Abstract

The present invention relates to the technical field of biogenic amine content detection. The present invention discloses a method for rapid scanning and detecting the content distribution of biogenic amines in fish meat, which comprises the following steps: obtaining a fluorescent image of fish meat to be tested, using a watershed algorithm to segment the grayscale fluorescent image to obtain multiple tissue regions; calculating the fat distribution index of each tissue region respectively, and taking the tissue region whose fat distribution index is greater than the average fat distribution index of all tissue regions as a high fat region; compensating the grayscale value of pixels in each high fat region in the grayscale fluorescent image to obtain a content distribution image; quantitatively marking the tissue region of fish meat where the fish meat tissue with a high biogenic amine concentration will be disturbed by the high fatness and the biogenic amine image will be a normal fluorescent color, so that the content distribution of the biogenic amines can be accurately identified by the naked eye or a quantitative evaluation system, reducing the interference of fat cells on the detection of the biogenic amine concentration, and obtaining an accurate content distribution image.

Description

Technical Field

[0001] The invention relates to the technical field of biogenic amine content detection, and in particular to a method for rapid scanning detection of fish meat biogenic amine content distribution. Background Art

[0002] At present, there are many analytical methods used for the detection of biogenic amines, such as thin layer chromatography, gas chromatography, liquid chromatography, etc. Among them, liquid chromatography is widely used, usually in conjunction with ultraviolet detectors, fluorescence detectors or mass spectrometers. These methods not only consume a lot of detection time, but also require a large amount of fish and extraction solvent in the extraction process, and the process is relatively cumbersome; the commonly used derivatization method is to use dansyl chloride and o-phthalaldehyde as derivatization reagents, the derivatization process requires heating and some derivatives are unstable.

[0003] In order to solve the defects of these existing technologies, a large number of studies on fluorescent probe methods have been conducted, which have realized the visual quantitative monitoring of biogenic amines and food freshness. For example, the Chinese invention patent with announcement number CN116082235B provides a DA-type organic fluorescent small molecule material and its preparation method, thin film fluorescent sensor and application, in which a set of quantitative evaluation systems based on ratiometric thin film fluorescent probes are realized, which realizes the visual quantitative monitoring of biogenic amines and food freshness, and can be combined with a smartphone to output RGB values ​​in real time for visual quantitative fluorescence detection of biogenic amines. Although this method has the characteristics of fast response speed, non-destructiveness and simple operation, its monitoring characteristics of biogenic amines gradually change from dark color (low biogenic amine concentration) to light color (high biogenic amine concentration) over time. In some fattened fish with high fat content, since histamine and its synthesis factors are mainly distributed in muscle tissue, although biogenic amines are also present in fat, the content of biogenic amines in muscle is higher than that in muscle. This is because muscle tissue is more active in the synthesis and accumulation of biogenic amines than other tissues, while adipose tissue may be relatively less in this regard. Therefore, the real-time distribution image of fish biogenic amines always shows that the area with more fish tissue has a high local concentration (the fluorescence color is relatively bright), while the area with more fat tissue has a low local concentration (the fluorescence color is relatively dark). However, if the fatness of the fish tissue is high, or the fat tissue accounts for the majority of the fish group, the fish tissue with a high concentration of biogenic amines will be disturbed by the fluorescent probe method, making the biogenic amine image appear as a normal fluorescence color, making it difficult to accurately identify the distribution of biogenic amines by the naked eye or quantitative evaluation system. Summary of the invention

[0004] The purpose of the present invention is to provide a method for rapid scanning and detecting the distribution of biogenic amine content in fish meat, so as to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0005] In order to achieve the above object, according to one aspect of the present invention, a method for rapid scanning detection of fish meat biogenic amine content distribution is provided, the method comprising the following steps:

[0006] S100, performing fluorescence treatment on the fish meat to be tested to obtain a fluorescence image of the fish meat to be tested;

[0007] S200, using the watershed algorithm to segment the grayscale fluorescence image to obtain multiple tissue regions;

[0008] S300, calculating the fertility distribution index of each tissue region respectively, and taking the tissue region whose fertility distribution index is greater than the average fertility distribution index of all tissue regions as a high fertility region;

[0009] S400, compensating the grayscale values ​​of pixels in each high-fertility area in the grayscale fluorescent image to obtain a content distribution image.

[0010] Furthermore, in S100, the method of performing fluorescence treatment on the fish meat to be tested is: coating the surface of the fish meat to be tested with fluorescent material and exciting the fluorescence with ultraviolet light. Preferably, the ultraviolet light is 365nm ultraviolet light.

[0011] Furthermore, the method for obtaining the fluorescent image of the fish meat to be tested is to take the fluorescent image of the fish meat to be tested by using a mobile phone camera of a smart phone.

[0012] Furthermore, the fluorescent material is a DA-type organic fluorescent small molecule material provided by the Chinese invention patent with publication number CN116082235B.

[0013] Preferably, in S100, the method for performing fluorescence treatment on the fish meat to be tested is: placing a thin film fluorescence sensor in the fish meat to be tested, and exciting fluorescence by ultraviolet rays.

[0014] The thin film fluorescence sensor is a thin film fluorescence sensor provided by the Chinese invention patent with publication number CN116082235B.

[0015] Preferably, the preparation method of the thin film fluorescence sensor is: mixing fluorescent material, polymethyl methacrylate and o-dichlorobenzene to obtain a mixed raw material; coating the mixed raw material on one side of a glass plate, and forming a fluorescent film on the surface of the glass plate after drying to obtain a thin film fluorescence sensor.

[0016] The volume ratio of the total mass of the fluorescent material and polymethyl methacrylate to the o-dichlorobenzene is 10-20 mg:1 mL; and the mass ratio of the fluorescent material to polymethyl methacrylate is 1:15.

[0017] Further, in S200, the method of using a watershed algorithm to segment the grayscale fluorescence image to obtain multiple tissue regions includes the following steps: identifying edge lines of the grayscale fluorescence image using a watershed algorithm, and then segmenting according to the edge lines to obtain multiple sub-images as tissue regions.

[0018] Since the muscle fat in each tissue area of ​​fish meat includes intramuscular fat inside the muscle and intermuscular fat between muscle bundles, the characteristics of intermuscular fat are obvious, so it is easier to identify the location of the area; while intramuscular fat is the fat deposited between muscle fibers (muscle cells) and muscle bundles (multiple muscle fibers are gathered into muscle bundles) in a muscle, which can be regarded as fat deposited in the entire muscle. Intramuscular fat has a greater impact on the accuracy of identifying biogenic amine concentrations. Since fat deposition in muscle tissue is not significant, the intramuscular fat area in the directly photographed content distribution map often shows a large difference from the imaging of ordinary muscles. The concentration growth rate gradient of biogenic amines produced in the intramuscular fat region is slower than that of ordinary muscle, but its concentration growth rate gradient is faster than that of intermuscular fat, which will be affected by the biogenic amines produced by muscle cells mixed in the intramuscular fat region. If the intramuscular fat region occupies a large area on the content distribution map, the intramuscular fat region is easily misidentified as an ordinary muscle region or an intermuscular fat region on the content distribution map, thereby affecting the observer or the recognition image to judge the freshness of the fish tissue as a whole. In order to accurately identify this high-fat intramuscular fat region, the present application provides the following method:

[0019] Further, in S300, the method of calculating the fertilizer distribution index of each tissue region is:

[0020] According to the acquisition order, each tissue region is sequentially formed into a first image sequence, and the average grayscale of the pixels of all tissue regions in the first image sequence is recorded as KM; the tissue regions whose average grayscale values ​​of all pixels are less than KM are screened from the first image sequence to form a second image sequence Pic, with i as the serial number, and pic(i) is the i-th tissue region in Pic; the grayscale fluorescence image is divided into multiple intervals by a grid division method (the grid size is set to 1 / 200 to 1 / 1000 of the fluorescence image size); (Since the higher biogenic amine concentration in the tissue region with more intramuscular fat is distributed in a dot-like or band-like manner in the tissue region, it is necessary to identify and determine whether the tissue region is an intramuscular fat region by dividing the interval);

[0021] Extract each tissue area into main distribution zones in turn, specifically:

[0022] In the range of i, the interval with the smallest average grayscale value among the intervals in the i-th tissue region pic(i) is Sgrid, and the interval with the largest average grayscale value among the intervals in the i-th tissue region pic(i) is Bgrid. (The bright interval Bgrid with the largest grayscale value is the interval area with the highest concentration of biogenic amines and more muscle cells, and the bright interval Sgrid with the smallest grayscale value is the interval area with the lowest concentration of biogenic amines and more fat. Biogenic amines will gradually diffuse from high-concentration areas to low-concentration areas. Note: There are generally one or more local high-concentration intervals on the main distribution belt of the intramuscular fat area. The main distribution belt is to mark the area where the fluctuation regularity of biogenic amines in the intramuscular fat area is more significant);

[0023] Take the point with the largest gray value in Bgrid as the anchor point P1, the point with the smallest gray value in Sgrid as the diffusion point P2, and the direction from P1 to P2 as the diffusion direction; draw a ray KL from point P1 to the diffusion direction; then all the intervals passed by KL are marked as diffusion intervals; according to the diffusion direction, all the diffusion intervals in pic(i) are sequentially formed into the main distribution belt starting from Bgrid;

[0024] Locate the diffusion limit interval in the main distribution zone, specifically:

[0025] j is the serial number of the diffusion interval in the main distribution zone; Gre(j) is the jth diffusion interval in the main distribution zone;

[0026] Judge each Gre(j) in the range of j in turn. If Gre(j) meets the diffusion limit condition, mark Gre(j) as the diffusion limit interval;

[0027] Among them, the diffusion limit condition is: SU(Gre(j-1))>SU(Gre(j)) and SU(Gre(j))<SU(Gre(j+1)), where SU(Gre(j)) is the diffusion rate, SU(Gre(j))=|GreD1-GreD2|÷|GreU1-GreU2|, where GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU2 is the grayscale value of the pixel with the largest grayscale value in the j+1th to Nth diffusion intervals; GreD1 is the pixel with the smallest grayscale value in the 1st to j-1th diffusion intervals; GreD2 is the grayscale value of the pixel with the smallest grayscale value in the j+1th to Nth diffusion intervals; N is the number of diffusion intervals in the main distribution band;

[0028] Preferably, the diffusion limit condition is: G(Gre(j-1))>G(Gre(j)) and G(Gre(j))<G(Gre(j+1)), where G(Gre(j)) is the grayscale value of all pixels in Gre(j);

[0029] The fertilizer distribution index of each tissue area is calculated according to the diffusion limit interval in the main distribution zone, specifically:

[0030] The serial number of the diffusion limit interval in the diffusion interval of the main distribution band is k; the absolute value of the difference between the average gray value of the diffusion interval from the 1st to the k-1th diffusion interval in the main distribution band and the average gray value of Bgrid is the diffusion surplus surplus; the absolute value of the difference between the average gray value of the diffusion interval from the k+1th to the Nth diffusion interval in the main distribution band and the average gray value of Sgrid is the diffusion deficit deficit; (the average gray value of the interval is the average gray value of all pixels in the interval, the diffusion surplus represents the intensity of the diffusion of the biogenic amine concentration from Bgrid, and the diffusion deficit represents the attenuation rate of the diffusion limit of the biogenic amine to Sgrid);

[0031] Traverse the diffusion limit intervals of each main distribution zone, and use BU to represent the number of diffusion limit intervals whose surplus corresponding to each diffusion limit interval is greater than the deficit; use BD to represent the number of diffusion limit intervals whose surplus corresponding to each diffusion limit interval is less than the deficit; then the calculation method of the fat distribution index FatIn is: FatIn=BU÷(BU+BD).

[0032] Among them, the fat distribution index determines the fatness of the fish tissue area by the ratio of the diffusion surplus of the diffusion of biogenic amines to the sum of the diffusion surplus and the diffusion deficit, thereby quantitatively marking the fish tissue area where the fish tissue with a higher concentration of biogenic amines will be disturbed by the high fatness and the biogenic amines will be imaged as a normal fluorescent color.

[0033] Since the volume of fat tissue of the same weight is 3 times larger than that of muscle tissue, and the intramuscular fat area is composed of muscle tissue and muscle cells, the latter of which has the function of contraction and movement, and the intramuscular fat area is fat tissue deposited in the whole muscle, therefore, the fat tissue in the intramuscular fat area squeezes the tissue in the surrounding area of ​​the intramuscular fat area due to problems such as the contraction of muscle cells, especially the tissue in the surrounding area is fluffy connective tissue or muscle tissue with a fast decay rate, and the intramuscular fat area has a relatively slow decay rate, so that the intramuscular fat area squeezes the surrounding area, so that the area of ​​the surrounding area is reduced, resulting in the tissue in the surrounding area The distribution map is not significant. In order to improve the significance of the tissue area of ​​the surrounding area, reduce the squeezing effect of the hydraulic pressure caused by the large volume of fat tissue on the tissue area of ​​the surrounding area, and improve the accuracy of positioning the intramuscular fat area, the present invention provides the following adjustment method:

[0034] Preferably, in S300, the range of each high fertility area needs to be adjusted, specifically:

[0035] The pixel with the largest gray value in the diffusion limit interval in the main distribution belt of the high fat area is AP; the pixel with the smallest gray value in the diffusion limit interval in the main distribution belt of the high fat area is BP; the point with the smallest difference in gray value with AP on the boundary of the high fat area is CP, and the point with the smallest difference in gray value with BP on the boundary of the high fat area is DP; the straight line passing through AP and CP is L1; the straight line passing through BP and DP is L2; ​​then the area between L1 and L2 is separated into two edge lines on the boundary of the high fat area, and the edge line with the highest average gray value of each point on the two edge lines is the muscle edge; (the muscle edge is not affected by the squeezing of fat tissue) the other edge line is the fat edge; the fat edge is deleted; the copied muscle edge is mirrored to obtain the mirror edge, and the mirror edge is spliced ​​at the position of the original fat edge, and the range adjustment of the high fat area is completed.

[0036] Through the above method, the diffusion characteristics of biogenic amines from high to low are utilized to mark the muscle edges with higher biogenic amine concentrations and the fat edges affected by the squeezing of adipose tissue. By mirroring the fat edges, the squeezing effect of adipose tissue in high-fat areas on the surrounding areas can be greatly avoided, thereby improving the characteristic significance of the tissue areas in the surrounding areas.

[0037] Further, in S400, the method of compensating the grayscale values ​​of pixels in each high fertility area in the grayscale fluorescent image to obtain the content distribution image is:

[0038] The maximum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM1; the minimum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM2;

[0039] The grayscale values ​​of pixels in each high fertility area are compensated in turn, specifically:

[0040] The diffusion limit interval corresponding to the pixel point with the largest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is APD, and the average gray value of the pixels in APD is AM; the diffusion limit interval corresponding to the pixel point with the smallest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is BPD, and the average gray value of the pixels in BPD is BM; the absolute value of the difference between GM1 and AM is DM1, the absolute value of the difference between GM2 and BM is DM2, and the minimum value between DM1 and DM2 is taken as the compensation difference; the gray value of each pixel in the high fertility area is increased by the compensation difference in sequence;

[0041] Obtain content distribution images.

[0042] Among them, the grayscale value corresponding to each pixel point on the content distribution image reflects the change in the content of biogenic amines in fish meat. By compensating the difference, it can greatly make up for the problem of low image recognition accuracy caused by the difference in the degradation of biogenic amines and the accumulation of organic compounds through amino acid degradation and amination of organic compounds in the fat tissue area during the deterioration process of fat tissue and muscle tissue in the intramuscular fat tissue area, thereby improving the data accuracy of biogenic amines.

[0043] Preferably, the method further comprises: mapping the content distribution image into a color image, specifically, by any one of a lookup table-based mapping method, a direct mapping method, and a pseudo-color processing method.

[0044] Preferably, the pseudo color processing method is: using the color mapping function in the OpenCV library to convert the content distribution image into a pseudo color image with a specific color effect.

[0045] Preferably, the biogenic amine content of the fish meat is inverted according to the grayscale value corresponding to each pixel point on the content distribution image.

[0046] The inversion method is to obtain the corresponding biogenic amine content by the gray value of each pixel according to the content mapping table.

[0047] Preferably, the method for obtaining the content mapping table is: using liquid chromatography to measure the biogenic amine content of fish meat at fixed intervals, and obtaining a content distribution image through steps S100 to S400 while collecting the biogenic amine content of the fish meat; recording the biogenic amine content corresponding to the grayscale value according to the content distribution image, and forming a content mapping table by the grayscale value and the corresponding biogenic amine content.

[0048] The beneficial effects of the present invention are as follows: the present invention provides a method for quickly scanning and detecting the distribution of biogenic amine content in fish meat, which quantitatively marks the tissue areas of fish meat where the fish meat tissue with higher biogenic amine concentration will be interfered by high fatness and the biogenic amine will be imaged as normal fluorescent color, so that the content distribution of biogenic amines can be accurately identified by the naked eye or a quantitative evaluation system, reducing the interference of fat cells in the detection of biogenic amine concentration and obtaining an accurate content distribution image. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0050] Figure 1 Shown is a flow chart of a method for rapid scanning detection of the distribution of biogenic amine content in fish meat. DETAILED DESCRIPTION

[0051] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0052] Example 1

[0053] like Figure 1 The flowchart of a method for rapid scanning and detecting the distribution of biogenic amine content in fish meat according to the present invention is shown below. Figure 1 To illustrate a rapid scanning detection method for the distribution of biogenic amine content in fish meat according to an embodiment of the present invention.

[0054] The specific steps include:

[0055] S100, performing fluorescence treatment on the fish meat to be tested to obtain a fluorescence image of the fish meat to be tested;

[0056] S200, using the watershed algorithm to segment the grayscale fluorescence image to obtain multiple tissue regions;

[0057] S300, calculating the fertility distribution index of each tissue region respectively, and taking the tissue region whose fertility distribution index is greater than the average fertility distribution index of all tissue regions as a high fertility region;

[0058] S400, compensating the grayscale values ​​of pixels in each high-fertility area in the grayscale fluorescent image to obtain a content distribution image.

[0059] Preferably, in S100, the method for performing fluorescence treatment on the fish meat to be tested is: placing a thin film fluorescence sensor in the fish meat to be tested, and exciting fluorescence by ultraviolet rays.

[0060] The thin film fluorescence sensor is a thin film fluorescence sensor provided by the Chinese invention patent with publication number CN116082235B.

[0061] Further, in S200, the method of using a watershed algorithm to segment the grayscale fluorescence image to obtain multiple tissue regions includes the following steps: identifying edge lines of the grayscale fluorescence image using a watershed algorithm, and then segmenting according to the edge lines to obtain multiple sub-images as tissue regions.

[0062] Further, in S300, the method of calculating the fertilizer distribution index of each tissue region is:

[0063] According to the acquisition order, each tissue area is sequentially formed into a first image sequence, and the average grayscale of the pixels of all tissue areas in the first image sequence is recorded as KM; the tissue areas with the average grayscale value of all pixels less than KM are screened from the first image sequence to form a second image sequence Pic, with i as the serial number, pic(i) is the i-th tissue area in Pic; the grayscale fluorescence image is divided into multiple intervals by a grid division method; wherein the grid size of the grid division method is set to 1 / 200 of the size of the fluorescence image.

[0064] Extract each tissue area into main distribution zones in turn, specifically:

[0065] In the range of i, the interval with the smallest average grayscale value among the intervals in the i-th tissue area pic(i) is Sgrid, and the interval with the largest average grayscale value among the intervals in the i-th tissue area pic(i) is Bgrid;

[0066] Take the point with the largest gray value in Bgrid as the anchor point P1, the point with the smallest gray value in Sgrid as the diffusion point P2, and the direction from P1 to P2 as the diffusion direction; draw a ray KL from point P1 to the diffusion direction; then all the intervals passed by KL are marked as diffusion intervals; according to the diffusion direction, all the diffusion intervals in pic(i) are sequentially formed into the main distribution belt starting from Bgrid;

[0067] Locate the diffusion limit interval in the main distribution zone, specifically:

[0068] j is the serial number of the diffusion interval in the main distribution zone; Gre(j) is the jth diffusion interval in the main distribution zone;

[0069] Judge each Gre(j) in the range of j in turn. If Gre(j) meets the diffusion limit condition, mark Gre(j) as the diffusion limit interval;

[0070] The diffusion limit condition is: G(Gre(j-1))>G(Gre(j)) and G(Gre(j))<G(Gre(j+1)), where G(Gre(j)) is the gray value of all pixels in Gre(j);

[0071] The fertilizer distribution index of each tissue area is calculated based on the diffusion limit interval in the main distribution zone, specifically:

[0072] The serial number of the diffusion limit interval in the diffusion interval of the main distribution band is recorded as k; the absolute value of the difference between the average gray value of the diffusion interval from the 1st to the k-1th diffusion interval in the main distribution band and the average gray value of Bgrid is the diffusion surplus; the absolute value of the difference between the average gray value of the diffusion interval from the k+1th to the Nth diffusion interval in the main distribution band and the average gray value of Sgrid is the diffusion deficit;

[0073] Traverse the diffusion limit intervals of each main distribution zone, and use BU to represent the number of diffusion limit intervals whose surplus corresponding to each diffusion limit interval is greater than the deficit; use BD to represent the number of diffusion limit intervals whose surplus corresponding to each diffusion limit interval is less than the deficit; then the calculation method of the fat distribution index FatIn is: FatIn=BU÷(BU+BD).

[0074] Further, in S400, the method of compensating the grayscale values ​​of pixels in each high fertility area in the grayscale fluorescent image to obtain the content distribution image is:

[0075] The maximum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM1; the minimum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM2;

[0076] The grayscale values ​​of pixels in each high fertility area are compensated in turn, specifically:

[0077] The diffusion limit interval corresponding to the pixel point with the largest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is APD, and the average gray value of the pixels in APD is AM; the diffusion limit interval corresponding to the pixel point with the smallest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is BPD, and the average gray value of the pixels in BPD is BM; the absolute value of the difference between GM1 and AM is DM1, the absolute value of the difference between GM2 and BM is DM2, and the minimum value between DM1 and DM2 is taken as the compensation difference; the gray value of each pixel in the high fertility area is increased by the compensation difference in sequence;

[0078] Obtain content distribution images.

[0079] Example 2

[0080] On the basis of Example 1, the diffusion limit condition of Example 1 is replaced by:

[0081] Among them, the diffusion limit conditions are: SU(Gre(j-1))>SU(Gre(j)) and SU(Gre(j))<SU(Gre(j+1)), where SU(Gre(j)) is the diffusion rate, SU(Gre(j))=|GreD1-GreD2|÷|GreU1-GreU2|, where GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU2 is the grayscale value of the pixel with the largest grayscale value in the j+1th to Nth diffusion intervals; GreD1 is the pixel with the smallest grayscale value in the 1st to j-1th diffusion intervals; GreD2 is the grayscale value of the pixel with the smallest grayscale value in the j+1th to Nth diffusion intervals; N is the number of diffusion intervals in the main distribution band.

[0082] Example 3

[0083] On the basis of Example 1, the following method is added in S300:

[0084] In S300, the range of each high fertility area also needs to be adjusted, specifically:

[0085] The pixel with the largest gray value in the diffusion limit interval in the main distribution belt of the high fat area is AP; the pixel with the smallest gray value in the diffusion limit interval in the main distribution belt of the high fat area is BP; the point with the smallest difference in gray value with AP on the boundary of the high fat area is CP, and the point with the smallest difference in gray value with BP on the boundary of the high fat area is DP; the straight line passing through AP and CP is L1; the straight line passing through BP and DP is L2; ​​then the area between L1 and L2 is separated into two edge lines on the boundary of the high fat area, and the edge line with the highest average gray value of each point on the two edge lines is the muscle edge; the other edge line is the fat edge; the fat edge is deleted; the copied muscle edge is mirrored to obtain the mirror edge, and the mirror edge is spliced ​​at the position of the original fat edge, and the range adjustment of the high fat area is completed.

[0086] Comparative Example 1, a DA-type organic fluorescent small molecule material and its preparation method, thin film fluorescent sensor and application thereof with publication number CN116082235B and the steps of Application Example 2 are used to obtain a fluorescent image of the fish meat to be tested as a content distribution image.

[0087] Live tilapia with a fatness greater than 4.6 are bled, slaughtered, scaled, and sliced ​​to obtain fish tissue blocks, and fish meat tissue with high intramuscular fat is artificially separated from the fish tissue blocks, washed in physiological saline, and the surface moisture is absorbed to obtain the fish meat to be tested.

[0088] Under an environment of 25° C., the grayscale content distribution image of the fish meat to be tested is measured every 4 hours according to the steps in the methods of Example 1, Example 2, Example 3 and Comparative Example 1 of the present invention, and the biogenic amine content of the fish meat is analyzed by liquid chromatography.

[0089] The biogenic amine content of the fish meat of Example 1, Example 2, Example 3 and Comparative Example 1 is inverted according to the grayscale value corresponding to each pixel point on the content distribution image of Example 1, Example 2, Example 3 and Comparative Example 1. The inversion method is to obtain the corresponding biogenic amine content according to the grayscale value of each pixel point according to the content mapping table.

[0090] Preferably, the method for obtaining the content mapping table is: using liquid chromatography to measure the biogenic amine content of fish meat every 4 hours after 8 hours, and obtaining a content distribution image while collecting the biogenic amine content of the fish meat; recording the biogenic amine content corresponding to the grayscale value according to the content distribution image, and forming a content mapping table by the grayscale value and the corresponding biogenic amine content.

[0091] SPSS software was used to analyze the Pearson correlation coefficient and root mean square error of the biogenic amine content of Example 1, Example 2, Example 3, Comparative Example 1 and the liquid chromatography method, and the analysis results were as follows:

[0092] The correlation coefficient between Example 1 and the liquid chromatography method was 0.8579, and the root mean square error was 1.5377;

[0093] The correlation coefficient between Example 2 and the liquid chromatography method was 0.9251, and the root mean square error was 1.3629;

[0094] The correlation coefficient between Example 3 and the liquid chromatography method was 0.9658, and the root mean square error was 1.2451;

[0095] The correlation coefficient between Comparative Example 1 and the liquid chromatography method is 0.4187, and the root mean square error is 2.3187.

[0096] From the analysis results of Example 1, Example 2, Example 3, Comparative Example 1 and the liquid chromatography method, in the fish tissue with intramuscular fat having a fatness greater than 4.6, the problem of low image recognition accuracy caused by the difference in the accumulation of biogenic amines through amino acid degradation and organic compound amination during the deterioration of the fat tissue and muscle tissue in the tissue area of ​​the intramuscular fat leads to a low correlation coefficient in Comparative Example 1. It can be seen that the scanning detection method proposed in Example 1, Example 2 and Example 3 of the present invention can be highly correlated with the measurement value of the liquid chromatography method.

[0097] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. A rapid scanning detection method for the distribution of biogenic amine content in fish meat, It is characterized in that The method comprises the following steps: S100, performing fluorescence treatment on the fish meat to be tested to obtain a fluorescence image of the fish meat to be tested; S200, using the watershed algorithm to segment the grayscale fluorescence image to obtain multiple tissue regions; S300, calculating the fertility distribution index of each tissue region respectively, and taking the tissue region whose fertility distribution index is greater than the average fertility distribution index of all tissue regions as a high fertility region; S400, compensating the gray value of the pixel in each high fertility area in the gray-scale fluorescence image to obtain a content distribution image; In S200, the method of using a watershed algorithm to segment a grayscale fluorescence image to obtain multiple tissue regions includes the following steps: identifying edge lines of the grayscale fluorescence image using a watershed algorithm, and then segmenting the grayscale fluorescence image according to the edge lines to obtain multiple sub-images as tissue regions respectively; In S300, the method for respectively calculating the fat distribution index of each tissue region is as follows: sequentially forming a first image sequence with each tissue region in the acquisition order, recording the average grayscale of the pixels of all tissue regions in the first image sequence as KM; screening the tissue regions whose average grayscale values ​​of all pixels are less than KM from the first image sequence to form a second image sequence Pic, with i as the sequence number, and pic(i) being the i-th tissue region in Pic; dividing the grayscale fluorescence image into a plurality of intervals by a grid division method; sequentially extracting each tissue region into a main distribution zone; locating the diffusion limit interval in the main distribution zone; and calculating the fat distribution index of each tissue region according to the diffusion limit interval in the main distribution zone; Among them, each tissue area is extracted into the main distribution zone in turn, specifically: In the range of i, the interval with the smallest average grayscale value among the intervals in the i-th tissue area pic(i) is Sgrid, and the interval with the largest average grayscale value among the intervals in the i-th tissue area pic(i) is Bgrid; Take the point with the largest gray value in Bgrid as the anchor point P1, the point with the smallest gray value in Sgrid as the diffusion point P2, and the direction from P1 to P2 as the diffusion direction; draw a ray KL from point P1 to the diffusion direction; then all the intervals passed by KL are marked as diffusion intervals; according to the diffusion direction, all the diffusion intervals in pic(i) are sequentially formed into the main distribution belt starting from Bgrid; Among them, the diffusion limit interval is located in the main distribution zone, specifically: j is the serial number of the diffusion interval in the main distribution zone; Gre(j) is the jth diffusion interval in the main distribution zone; each Gre(j) is judged in turn within the range of j, and if Gre(j) meets the diffusion limit condition, Gre(j) is marked as the diffusion limit interval; Among them, the fertilizer distribution index of each tissue area is calculated according to the diffusion limit interval in the main distribution zone, specifically: The serial number of the diffusion limit interval in the diffusion interval of the main distribution band is recorded as k; the absolute value of the difference between the average gray value of the diffusion interval from the 1st to the k-1th diffusion interval in the main distribution band and the average gray value of Bgrid is the diffusion surplus; the absolute value of the difference between the average gray value of the diffusion interval from the k+1th to the Nth diffusion interval in the main distribution band and the average gray value of Sgrid is the diffusion deficit; Traverse the diffusion limit intervals of each main distribution zone, and use BU to represent the number of diffusion limit intervals whose surplus is greater than the deficit corresponding to each diffusion limit interval; use BD to represent the number of diffusion limit intervals whose surplus is less than the deficit corresponding to each diffusion limit interval; then the calculation method of the fertilizer distribution index FatIn is: FatIn=BU÷(BU+BD); In S400, the method for compensating the grayscale values ​​of pixels in each high-fertility area in the grayscale fluorescent image to obtain a content distribution image is: The maximum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM1; the minimum gray value of the pixels in each high fertility area in the gray-scale fluorescent image is GM2; The grayscale values ​​of pixels in each high fertility area are compensated in turn, specifically: The diffusion limit interval corresponding to the pixel point with the largest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is APD, and the average gray value of the pixels in APD is AM; the diffusion limit interval corresponding to the pixel point with the smallest gray value in the diffusion limit interval of the main distribution belt of the high fertility area is BPD, and the average gray value of the pixels in BPD is BM; the absolute value of the difference between GM1 and AM is DM1, the absolute value of the difference between GM2 and BM is DM2, and the minimum value between DM1 and DM2 is taken as the compensation difference; the gray value of each pixel in the high fertility area is increased by the compensation difference in sequence; Obtain content distribution images.

2. A method for rapid scanning and detecting the distribution of biogenic amine content in fish meat according to claim 1, It is characterized in that The diffusion limit condition is: SU(Gre(j-1))>SU(Gre(j)) and SU(Gre(j))<SU(Gre(j+1)), where SU(Gre(j)) is the diffusion rate, SU(Gre(j))=|GreD1-GreD2|÷|GreU1-GreU2|, where GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU1 is the grayscale value of the pixel with the largest grayscale value in the 1st to j-1th diffusion intervals; GreU2 is the grayscale value of the pixel with the largest grayscale value in the j+1th to Nth diffusion intervals; GreD1 is the pixel with the smallest grayscale value in the 1st to j-1th diffusion intervals; GreD2 is the grayscale value of the pixel with the smallest grayscale value in the j+1th to Nth diffusion intervals; N is the number of diffusion intervals in the main distribution band.

3. A method for rapid scanning and detecting the distribution of biogenic amine content in fish meat according to claim 1, It is characterized in that The diffusion limit condition is: G(Gre(j-1))>G(Gre(j)) and G(Gre(j))<G(Gre(j+1)), where G(Gre(j)) is the grayscale value of all pixels in Gre(j).

4. A method for rapid scanning and detecting the distribution of biogenic amine content in fish meat according to claim 1, It is characterized in that In S300, the range of each high fertility area needs to be adjusted, specifically: the pixel point with the largest gray value in the diffusion limit interval in the main distribution zone of the high fertility area is recorded as AP; the pixel point with the smallest gray value in the diffusion limit interval in the main distribution zone of the high fertility area is recorded as BP; the point with the smallest gray value difference with AP on the boundary of the high fertility area is recorded as CP, and the point with the smallest gray value difference with BP on the boundary of the high fertility area is recorded as DP; the straight line passing through AP and CP is recorded as L1; The straight line passing through BP and DP is L2; ​​then the area between L1 and L2 is separated into two edge lines on the boundary of the high fat area, and the edge line with the highest average gray value of each point on the two edge lines is the muscle edge; The other edge line is the fat edge; Delete the fat edge; mirror the copied muscle edge to obtain a mirror edge, and splice the mirror edge to the position of the original fat edge. The range adjustment of the high fat area is completed.

Images