An image analysis-based method for comparing the virulence of aeromonas salmonicida

Abstract

The application discloses a kind of based on image analysis's Aeromonas salmonicida virulence comparison method, this method includes: obtaining Aeromonas salmonicida sample, carries out image acquisition, obtains the image data of Aeromonas salmonicida;Image data is preprocessed and Canny edge detection, and the microorganism image is segmented, and target microorganism area is extracted;From the image after segmentation, extract the characteristic related to Aeromonas salmonicida virulence;The extracted feature is compared with known Aeromonas salmonicida virulence factor database, identifies the known virulence factor most similar to the feature of the bacteria to be measured, assesses the virulence intensity of the bacteria to be measured;The data generated in the comparison process is sorted and summarized, and the comparison and evaluation results are output in a visual form.The method of the application can quickly and accurately compare the virulence characteristics of Aeromonas salmonicida from different sources or under different conditions, and can more intuitively reflect the differences between strains, thereby improving the accuracy and efficiency of virulence comparison.

Description

Technical Field

[0001] This invention relates to the field of computer image processing and analysis technology, and in particular to a method for comparing the virulence of Aeromonas salmonidae based on image analysis. Background Technology

[0002] Aeromonas salmonid is a bacterium widely found in aquatic environments, pathogenic to a variety of aquatic animals, and causing significant economic losses to the aquaculture industry. Traditionally, the virulence assessment and comparison of Aeromonas salmonid has relied primarily on laboratory biochemical tests, a method that is not only time-consuming and labor-intensive but also difficult to perform rapid and accurate analysis of large numbers of samples.

[0003] With the rapid development of computer technology and image processing technology, image analysis is increasingly widely used in biological and medical research. Through image analysis, morphological and structural features of microorganisms can be extracted, enabling rapid identification and classification. This method has advantages such as high efficiency, accuracy, and non-contact processing, making it particularly suitable for processing and analyzing large-scale samples.

[0004] Therefore, this invention proposes an image analysis-based method for comparing the virulence of Aeromonas salmonicidae, aiming to accurately compare and analyze the virulence of Aeromonas salmonicidae from different sources or under different conditions through advanced image recognition, processing and machine learning technologies. Summary of the Invention

[0005] The purpose of this invention is to propose an image analysis-based method for comparing the virulence of Aeromonas salmonii, which addresses the problems of time-consuming, labor-intensive, and difficult-to-accurate analysis of large numbers of samples in existing technologies.

[0006] To achieve the above objectives, the present invention employs the following technical solution: a method for comparing the virulence of Aeromonas salmonicida based on image analysis, comprising the following steps:

[0007] Step S1: Obtain Aeromonas salmonicida samples from different sources, and acquire images of the samples using a high-resolution camera to obtain image data of Aeromonas salmonicida.

[0008] Step S2: Preprocess the acquired image data and perform Canny edge detection, and segment the microbial image through morphological processing to extract the target microbial region;

[0009] Step S3: Extract features related to the virulence of Aeromonas salmonicida from the segmented images using image analysis technology;

[0010] Step S4: Compare the extracted features with a database of known Aeromonas salmonicida virulence factors to identify the known virulence factors that are most similar to the characteristics of the bacteria to be tested, and evaluate the virulence of the bacteria to be tested by combining the correspondence in the database.

[0011] Step S5: Organize and summarize the data generated during the comparison process, compile a comparison report, and output the comparison and evaluation results in a visual form.

[0012] Furthermore, step S1 also includes the following sub-steps:

[0013] S1-1, Collect samples containing Aeromonas salmonicidae from different environments or hosts;

[0014] S1-2, Aeromonas salmonidae was isolated from the collected samples using the streak stripping method, and the purity and activity of the bacteria were ensured through multiple subcultures.

[0015] S1-3, Select a high-resolution camera as a Canon EOS 80D camera, with an ISO speed of 800, an exposure time of 1 / 200s, and an image resolution of 4000×6000;

[0016] S1-4, Place Aeromonas salmonidae in the observation area and perform imaging operations to obtain high-resolution bacterial cell image data.

[0017] Furthermore, step S2 also includes the following sub-steps:

[0018] S2-1 converts the original bacterial cell image into a grayscale image, enhances contrast through point operations, and removes noise through Gaussian filtering.

[0019] S2-2, Let f(x,y) be the grayscale function in the bacterial cell image. Calculate the gradient ∇f, the magnitude of the gradient, and the gradient direction at each pixel (x,y). The specific formula is as follows:

[0020]

[0021]

[0022]

[0023] in, and These are the gradient components along the x-axis and y-axis, respectively, and their magnitudes. Indicates the strength of the gradient. Indicates the direction of the gradient;

[0024] S2-3, Non-maximum suppression is performed on the gradient magnitude in the four gradient directions of 0°, 45°, 90° and 135°. For each pixel (x, y), two intersection points are found in its 8-neighborhood according to the gradient direction. If the gradient magnitude of the center point is the maximum value among the three points, the point is retained; otherwise, it is set to 0.

[0025] S2-4: Set a low threshold and a high threshold. If the gradient magnitude of a pixel is less than the low threshold, discard the pixel. If the gradient magnitude is greater than the high threshold, retain the pixel as a strong edge point. If the gradient magnitude is between the two, check the pixels in its 8-neighborhood. If the gradient magnitude of any point in the neighborhood is greater than the high threshold, retain the pixel as a weak edge point. Otherwise, discard the pixel.

[0026] S2-5, perform morphological processing on the image after edge detection, close the edge image to obtain the segmentation result, the morphological processing includes dilation, erosion and closing operations.

[0027] Furthermore, step S3 also includes the following sub-steps:

[0028] S3-1, Extract features related to the virulence of Aeromonas salmonicida from the segmented image, wherein the virulence-related features of Aeromonas salmonicida include geometric features, internal structure histogram features, and Fourier descriptors;

[0029] S3-2, by measuring and recording the geometric characteristics of Aeromonas salmonidae, including the area, perimeter, aspect ratio, and circularity of the bacterial cell;

[0030] S3-3, mark K sampling points at equal intervals on the bacterial cell boundary outline. For any three sampling points, calculate the inscribed angle between them. Count all the obtainable inscribed angles according to a certain angle range to obtain the internal structure histogram features.

[0031] S3-4, mark M sampling points at equal intervals on the bacterial cell boundary outline, calculate the Euclidean distance from each sampling point to the center point, and arrange them in order as { , ,..., The distance sequence is then subjected to a Discrete Fourier Transform to extract the frequency domain components as Fourier descriptors. The specific formula is as follows:

[0032]

[0033] in, Let represent the s-th complex coefficient, s = 0, 1, ..., M-1; j represents the unit imaginary part; t represents the Euclidean distance from the t-th sampling point to the center point; Standardization: Internal standardization is performed on the Fourier coefficients to reduce the effects of noise, translation, and rotation. The specific formula is: Since the boundary contour of microorganisms can be represented as a real function, It has conjugate symmetry, and finally we get a set 3D feature vectors ( ,..., ).

[0034] Furthermore, step S4 also includes the following sub-steps:

[0035] S4-1, compare the extracted bacterial cell feature vectors with the feature vectors in the virulence factor database one by one, and find the most similar feature vector pair using the nearest neighbor algorithm. The specific formula is as follows:

[0036]

[0037] Where p and q are two n-dimensional vectors, and These are the i-th components of vectors p and q, respectively;

[0038] S4-2 calculates the similarity value of each pair of matched feature vectors using cosine similarity. The higher the similarity value, the more similar the two feature vectors are, and thus they have similar toxicity characteristics. The specific formula is as follows:

[0039]

[0040] Where A*B is the dot product of vectors A and B. and These are the magnitudes of vectors A and B, respectively;

[0041] S4-3, based on the matched feature vectors and similarity values, identify the known virulence factors that are most similar to the characteristics of the bacterial cells to be tested;

[0042] S4-4. Based on the identified virulence factors and the correspondence between virulence factors and virulence strength in the virulence factor database, assess the virulence strength of the test bacteria.

[0043] Furthermore, step S5 also includes the following sub-steps:

[0044] S5-1, Organize and summarize all the data generated during the comparison process. The data includes the extracted feature vectors, similarity calculation results, virulence factor identification results, and virulence strength assessment values.

[0045] S5-2, Generate charts based on the collected data, including scatter plots, bar charts, line charts, and heatmaps;

[0046] S5-3, Compile a comparison report and embed the generated charts into the report. The report includes a description of the comparison method, the data processing steps, and an analysis and discussion of the comparison results.

[0047] The beneficial effects of the technical solution provided by this invention include at least the following:

[0048] This invention obtains Aeromonas salmonicida samples from various sources, acquires images of the samples using a high-resolution camera to obtain image data of Aeromonas salmonicida, preprocesses the acquired image data and performs Canny edge detection, and segments the microbial images through morphological processing to extract target microbial regions; extracts virulence-related features of Aeromonas salmonicida from the segmented images using image analysis technology; compares the extracted features with a known database of Aeromonas salmonicida virulence factors to identify the known virulence factors most similar to the characteristics of the target bacterial cell, and evaluates the virulence strength of the target bacterial cell based on the correspondence in the database; organizes and summarizes the data generated during the comparison process, compiles a comparison report, and outputs the comparison and evaluation results in a visual form.

[0049] This invention utilizes image analysis technology to quickly and accurately compare the virulence characteristics of Aeromonas salmonidae from different sources or under different conditions, providing a more intuitive reflection of the differences between strains and thus improving the accuracy and efficiency of virulence comparison.

[0050] This method utilizes advanced image recognition and processing technologies, as well as machine learning algorithms, to achieve automated analysis and comparison of Aeromonas salmonidae images. This not only reduces the burden of manual operation but also improves the accuracy and consistency of the analysis.

[0051] This method enables rapid identification and virulence comparison of Aeromonas salmonidae, providing a scientific basis for disease control and prevention, helping to reduce the occurrence and spread of diseases, improve the production efficiency and product quality of aquaculture, and thus promote the healthy development of the aquaculture industry. Attached Figure Description

[0052] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart of a method provided in an embodiment of the present invention. Detailed Implementation

[0054] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an image analysis-based method for comparing the virulence of Aeromonas salmonis according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0056] The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0057] The following describes in detail, with reference to the accompanying drawings, a specific scheme for comparing the virulence of Aeromonas salmonidae based on image analysis provided by the present invention.

[0058] Example

[0059] Please see Figure 1 The diagram illustrates a flowchart of an image analysis-based method for comparing the virulence of Aeromonas salmonii, according to an embodiment of the present invention. The method includes the following steps:

[0060] Step S1: Obtain Aeromonas salmonicida samples from different sources, and acquire images of the samples using a high-resolution camera to obtain image data of Aeromonas salmonicida.

[0061] Step S1 further includes the following sub-steps:

[0062] S1-1, Collect samples containing Aeromonas salmonicidae from different environments or hosts;

[0063] S1-2, Aeromonas salmonidae was isolated from the collected samples using the streak stripping method, and the purity and activity of the bacteria were ensured through multiple subcultures.

[0064] S1-3, Select a high-resolution camera as a Canon EOS 80D camera, with an ISO speed of 800, an exposure time of 1 / 200s, and an image resolution of 4000×6000;

[0065] S1-4, Place Aeromonas salmonidae in the observation area and perform imaging operations to obtain high-resolution bacterial cell image data.

[0066] It should be noted that Aeromonas salmonidae exists in a variety of environments and hosts, including freshwater environments, marine environments, and in fish or other aquatic organisms. Sample collection needs to cover these different sources to ensure sample diversity and representativeness.

[0067] During the collection process, the samples are preserved in appropriate preservation solutions to avoid contamination and deterioration, ensuring the integrity and activity of the samples, and are transported to the laboratory for further processing as soon as possible.

[0068] Streak plating is a commonly used method for microbial isolation. It involves streaking a single colony onto a culture medium to isolate a pure strain of Aeromonas salmonidae. To ensure the purity and activity of the bacteria, multiple subcultures are performed, which not only further purifies the bacteria but also enhances their adaptability and reproductive capacity.

[0069] The Canon EOS 80D is a high-performance, high-resolution camera suitable for capturing images of microorganisms. ISO speed, exposure time, and image resolution are key factors affecting image quality. An ISO speed of 800 can reduce noise while ensuring image sharpness. An exposure time of 1 / 200s ensures that the image will not be blurred due to overexposure. An image resolution of 4000×6000 provides sufficient detail and resolution.

[0070] The observation area setup includes placing Aeromonas salmonidae in an appropriate observation area to ensure that the bacteria are clearly visible; the imaging operation involves adjusting the camera parameters to image the bacteria.

[0071] Step S2: Preprocess the acquired image data and perform Canny edge detection, and segment the microbial image through morphological processing to extract the target microbial region;

[0072] Step S2 further includes the following sub-steps:

[0073] S2-1 converts the original bacterial cell image into a grayscale image, enhances contrast through point operations, and removes noise through Gaussian filtering.

[0074] S2-2, Let f(x,y) be the grayscale function in the bacterial cell image. Calculate the gradient ∇f, the magnitude of the gradient, and the gradient direction at each pixel (x,y). The specific formula is as follows:

[0075]

[0076]

[0077]

[0078] in, and These are the gradient components along the x-axis and y-axis, respectively, and their magnitudes. Indicates the strength of the gradient. Indicates the direction of the gradient;

[0079] S2-3, Non-maximum suppression is performed on the gradient magnitude in the four gradient directions of 0°, 45°, 90° and 135°. For each pixel (x, y), two intersection points are found in its 8-neighborhood according to the gradient direction. If the gradient magnitude of the center point is the maximum value among the three points, the point is retained; otherwise, it is set to 0.

[0080] S2-4: Set a low threshold and a high threshold. If the gradient magnitude of a pixel is less than the low threshold, discard the pixel. If the gradient magnitude is greater than the high threshold, retain the pixel as a strong edge point. If the gradient magnitude is between the two, check the pixels in its 8-neighborhood. If the gradient magnitude of any point in the neighborhood is greater than the high threshold, retain the pixel as a weak edge point. Otherwise, discard the pixel.

[0081] S2-5, perform morphological processing on the image after edge detection, close the edge image to obtain the segmentation result, the morphological processing includes dilation, erosion and closing operations.

[0082] It should be noted that grayscale image conversion converts the original bacterial cell image from color to grayscale. Grayscale images only contain brightness information, which can simplify subsequent processing steps. Contrast enhancement enhances the contrast of the image through point operations (linear transformation), making the difference between the target microorganism and the background more obvious.

[0083] Gaussian filtering denoising is a process that uses a Gaussian filter to smooth an image, thereby reducing the impact of noise on subsequent processing. A Gaussian filter is a linear filter whose weighting coefficients are determined by a Gaussian function, and it can effectively remove high-frequency noise from an image.

[0084] The gradient is a vector that represents the rate of change and direction of an image at a given point; the gradient magnitude represents the strength of the gradient, i.e. how fast the image changes at that point; and the gradient direction represents the direction of change.

[0085] Morphological processing of the image after edge detection can improve the continuity of the edges, fill small holes, and remove small noise points. Dilation can expand the edge region, erosion can shrink the edge region, and closing operation is to first dilate and then erode, which helps to close the edge image and obtain the final segmentation result.

[0086] Step S3: Extract features related to the virulence of Aeromonas salmonicida from the segmented images using image analysis techniques;

[0087] Step S3 further includes the following sub-steps:

[0088] S3-1, Extract features related to the virulence of Aeromonas salmonicida from the segmented image, wherein the virulence-related features of Aeromonas salmonicida include geometric features, internal structure histogram features, and Fourier descriptors;

[0089] S3-2, by measuring and recording the geometric characteristics of Aeromonas salmonidae, including the area, perimeter, aspect ratio, and circularity of the bacterial cell;

[0090] S3-3, mark K sampling points at equal intervals on the bacterial cell boundary outline. For any three sampling points, calculate the inscribed angle between them. Count all the obtainable inscribed angles according to a certain angle range to obtain the internal structure histogram features.

[0091] S3-4, mark M sampling points at equal intervals on the bacterial cell boundary outline, calculate the Euclidean distance from each sampling point to the center point, and arrange them in order as { , ,..., The distance sequence is then subjected to a Discrete Fourier Transform to extract the frequency domain components as Fourier descriptors. The specific formula is as follows:

[0092]

[0093] in, Let represent the s-th complex coefficient, s = 0, 1, ..., M-1; j represents the unit imaginary part; t represents the Euclidean distance from the t-th sampling point to the center point; Standardization: Internal standardization is performed on the Fourier coefficients to reduce the effects of noise, translation, and rotation. The specific formula is: Since the boundary contour of microorganisms can be represented as a real function, It has conjugate symmetry, and finally we get a set 3D feature vectors ( ,..., ).

[0094] It should be noted that geometric features describe the basic shape and size information of microorganisms, which helps to distinguish different morphologies of bacteria and thus associate them with virulence. Area: the number of pixels or area occupied by the bacterial cell; perimeter: the total length of the bacterial cell's boundary outline; aspect ratio: the ratio of the longest axis to the shortest axis of the bacterial cell, reflecting the shape of the bacterial cell; roundness: describing how close the shape of the bacterial cell is to a circle.

[0095] The internal structure histogram feature can obtain the statistical characteristics of the internal structure of bacteria, which can be used to describe the internal spatial structure of an image, reflect the distribution of pixel intensity in the image, and reflect the complexity and symmetry of the internal structure of bacteria, thus being related to virulence.

[0096] Fourier descriptors are a frequency domain-based feature extraction method that uses frequency domain information to distinguish different contours and thus identify objects. They can describe the shape and complexity of bacterial cell boundary contours and reflect the periodic changes and irregularities of bacterial cell boundaries.

[0097] Step S4: Compare the extracted features with the known Aeromonas salmonicida virulence factor database, identify the known virulence factors that are most similar to the characteristics of the bacteria to be tested, and evaluate the virulence of the bacteria to be tested by combining the correspondence in the database.

[0098] S4-1, compare the extracted bacterial cell feature vectors with the feature vectors in the virulence factor database one by one, and find the most similar feature vector pair using the nearest neighbor algorithm. The specific formula is as follows:

[0099]

[0100] Where p and q are two n-dimensional vectors, and These are the i-th components of vectors p and q, respectively;

[0101] S4-2 calculates the similarity value of each pair of matched feature vectors using cosine similarity. The higher the similarity value, the more similar the two feature vectors are, and thus they have similar toxicity characteristics. The specific formula is as follows:

[0102]

[0103] Where A*B is the dot product of vectors A and B. and These are the magnitudes of vectors A and B, respectively;

[0104] S4-3, based on the matched feature vectors and similarity values, identify the known virulence factors that are most similar to the characteristics of the bacterial cells to be tested;

[0105] S4-4. Based on the identified virulence factors and the correspondence between virulence factors and virulence strength in the virulence factor database, assess the virulence strength of the test bacteria.

[0106] It should be noted that the nearest neighbor algorithm is a commonly used search algorithm used to find the object most similar to the query object in a database. It can find the virulence factor feature vector that is most similar to the feature vector of the bacterial cell to be tested. In specific implementation, Euclidean distance can be used to calculate the distance between two feature vectors, and the feature vector pair with the smallest distance is selected as the most similar feature vector pair.

[0107] Cosine similarity is a measure of the similarity between the directions of two vectors. Its value range is [-1, 1]. When the two vectors are in the same direction, the cosine similarity is 1; when the two vectors are in opposite directions, the cosine similarity is -1; and when the two vectors are perpendicular, the cosine similarity is 0.

[0108] Based on the calculated similarity values, the known virulence factor with the highest similarity value is selected as the virulence factor most similar to the characteristics of the bacterial cell to be tested. If multiple virulence factors have high similarity values, the similarity values ​​can be sorted, and the virulence factor corresponding to the highest value or the top N values ​​after sorting can be selected.

[0109] The virulence factor database contains a correspondence between each virulence factor and its virulence strength. The correspondence can be quantitative (virulence index) or qualitative (e.g., strong, moderate, weak). Based on the identified virulence factors and their correspondence in the database, the virulence strength of the test bacteria can be assessed. If the identified virulence factor has high virulence in the database, the test bacteria are highly virulent; if the identified virulence factor has weak virulence in the database, the test bacteria are weakly virulent.

[0110] Step S5: Organize and summarize the data generated during the comparison process, compile a comparison report, and output the comparison and evaluation results in a visual form;

[0111] Step S5 further includes the following sub-steps:

[0112] S5-1, Organize and summarize all the data generated during the comparison process. The data includes the extracted feature vectors, similarity calculation results, virulence factor identification results, and virulence strength assessment values.

[0113] S5-2, Generate charts based on the collected data, including scatter plots, bar charts, line charts, and heatmaps;

[0114] S5-3, Compile a comparison report and embed the generated charts into the report. The report includes a description of the comparison method, the data processing steps, and an analysis and discussion of the comparison results.

[0115] It should be noted that the extracted feature vector is the feature information extracted from the original data for comparison; the similarity calculation result is the numerical result that measures the degree of similarity between different samples or sequences; the virulence factor identification result is the specific factor or feature related to virulence identified through comparison and analysis; and the virulence strength assessment value is the numerical value that quantifies the virulence strength of the sample or sequence based on the virulence factor identification result.

[0116] Scatter plot: used to show the relationship between two variables, such as the correlation between feature vectors; Bar chart: used to show the comparison of data in different categories or groups, such as the virulence strength assessment values ​​of different samples; Line chart: used to show the trend of data over time or other continuous variables, such as the change of virulence factor identification results with the comparison process; Heat map: used to show the complex relationship between multiple variables, such as the similarity matrix between feature vectors.

[0117] In this way, an image analysis-based method for comparing the virulence of Aeromonas salmonicida can intuitively display the virulence characteristics and differences of Aeromonas salmonicida from different sources.

[0118] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for comparing the virulence of Aeromonas salmonicida based on image analysis, characterized in that, The method includes: Step S1: Obtain Aeromonas salmonicida samples from different sources, and acquire images of the samples using a high-resolution camera to obtain image data of Aeromonas salmonicida. Step S2: Preprocess the acquired image data and perform Canny edge detection, and segment the microbial image through morphological processing to extract the target microbial region; Step S3: Extract features related to the virulence of Aeromonas salmonicida from the segmented images using image analysis technology; Step S4: Compare the extracted features with a database of known Aeromonas salmonicida virulence factors to identify the known virulence factors that are most similar to the characteristics of the bacteria to be tested, and evaluate the virulence of the bacteria to be tested by combining the correspondence in the database. Step S5: Organize and summarize the data generated during the comparison process, compile a comparison report, and output the comparison and evaluation results in a visual form.

2. The image analysis-based virulence comparison method for Aeromonas salmonidae as described in claim 1, characterized in that: Step S1 further includes the following sub-steps: S1-1, Collect samples containing Aeromonas salmonicidae from different environments or hosts; S1-2, Aeromonas salmonidae was isolated from the collected samples using the streak stripping method, and the purity and activity of the bacteria were ensured through multiple subcultures. S1-3, Select a high-resolution camera as a Canon EOS 80D camera, with an ISO speed of 800, an exposure time of 1 / 200s, and an image resolution of 4000×6000; S1-4, Place Aeromonas salmonidae in the observation area and perform imaging operations to obtain high-resolution bacterial cell image data.

3. The image analysis-based virulence comparison method for Aeromonas salmonidae as described in claim 1, characterized in that: Step S2 further includes the following sub-steps: S2-1 converts the original bacterial cell image into a grayscale image, enhances contrast through point operations, and removes noise through Gaussian filtering. S2-2, Let f(x,y) be the grayscale function in the bacterial cell image. Calculate the gradient ∇f, the magnitude of the gradient, and the gradient direction at each pixel (x,y). The specific formula is as follows: ; ; ; in, and These are the gradient components along the x-axis and y-axis, respectively, and their magnitudes. Indicates the strength of the gradient. Indicates the direction of the gradient; S2-3, Non-maximum suppression is performed on the gradient magnitude in the four gradient directions of 0°, 45°, 90° and 135°. For each pixel (x, y), two intersection points are found in its 8-neighborhood according to the gradient direction. If the gradient magnitude of the center point is the maximum value among the three points, the point is retained; otherwise, it is set to 0. S2-4: Set a low threshold and a high threshold. If the gradient magnitude of a pixel is less than the low threshold, discard the pixel. If the gradient magnitude is greater than the high threshold, retain the pixel as a strong edge point. If the gradient magnitude is between the two, check the pixels in its 8-neighborhood. If the gradient magnitude of any point in the neighborhood is greater than the high threshold, retain the pixel as a weak edge point. Otherwise, discard the pixel. S2-5, perform morphological processing on the image after edge detection, close the edge image to obtain the segmentation result, the morphological processing includes dilation, erosion and closing operations.

4. The image analysis-based virulence comparison method for Aeromonas salmonidae as described in claim 1, characterized in that: Step S3 further includes the following sub-steps: S3-1, Extract features related to the virulence of Aeromonas salmonicida from the segmented image, wherein the virulence-related features of Aeromonas salmonicida include geometric features, internal structure histogram features, and Fourier descriptors; S3-2, by measuring and recording the geometric characteristics of Aeromonas salmonidae, including the area, perimeter, aspect ratio, and circularity of the bacterial cell; S3-3, mark K sampling points at equal intervals on the bacterial cell boundary outline. For any three sampling points, calculate the inscribed angle between them. Count all the obtainable inscribed angles according to a certain angle range to obtain the internal structure histogram features. S3-4, mark M sampling points at equal intervals on the bacterial cell boundary outline, calculate the Euclidean distance from each sampling point to the center point, and arrange them in order as { , ,..., The distance sequence is then subjected to a Discrete Fourier Transform to extract the frequency domain components as Fourier descriptors. The specific formula is as follows: ; in, Let represent the s-th complex coefficient, s = 0, 1, ..., M-1; j represents the unit imaginary part; t represents the Euclidean distance from the t-th sampling point to the center point; Standardization: Internal standardization is performed on the Fourier coefficients to reduce the effects of noise, translation, and rotation. The specific formula is: Since the boundary contour of microorganisms can be represented as a real function, It has conjugate symmetry, and finally we get a set 3D feature vectors ( ,..., ).

5. The image analysis-based virulence comparison method for Aeromonas salmonicidae as described in claim 1, characterized in that: Step S4 further includes the following sub-steps: S4-1, compare the extracted bacterial cell feature vectors with the feature vectors in the virulence factor database one by one, and find the most similar feature vector pair using the nearest neighbor algorithm. The specific formula is as follows: ; Where p and q are two n-dimensional vectors, and These are the i-th components of vectors p and q, respectively; S4-2 calculates the similarity value of each pair of matched feature vectors using cosine similarity. The higher the similarity value, the more similar the two feature vectors are, and thus they have similar toxicity characteristics. The specific formula is as follows: ; Where A*B is the dot product of vectors A and B. and These are the magnitudes of vectors A and B, respectively; S4-3, based on the matched feature vectors and similarity values, identify the known virulence factors that are most similar to the characteristics of the bacterial cells to be tested; S4-4. Based on the identified virulence factors and the correspondence between virulence factors and virulence strength in the virulence factor database, assess the virulence strength of the test bacteria.

6. The method for comparing the virulence of Aeromonas salmonicida based on image analysis as described in claim 1, characterized in that: Step S5 further includes the following sub-steps: S5-1, Organize and summarize all the data generated during the comparison process. The data includes the extracted feature vectors, similarity calculation results, virulence factor identification results, and virulence strength assessment values. S5-2, Generate charts based on the collected data, including scatter plots, bar charts, line charts, and heatmaps; S5-3, Compile a comparison report and embed the generated charts into the report. The report includes a description of the comparison method, the data processing steps, and an analysis and discussion of the comparison results.

Images