Method and system for improving incubation management efficiency of egg-carrying shrimps

Abstract

The invention relates to the technical field of aquaculture management, in particular to a method and system for improving incubation management efficiency of egg-carrying shrimps. Performing feature extraction and de-noising processing on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; performing registration processing on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model; constructing a morphological feature three-dimensional model diagram of the target cherax quadricarinatus according to the feature point three-dimensional model of the target cherax quadricarinatus; importing the morphological characteristic three-dimensional model diagram of the target cherax quadricarinatus into a knowledge base to estimate the brood amount of the target cherax quadricarinatus to obtain the brood amount of the target cherax quadricarinatus; and classifying and screening the target cherax quadricarinatus according to the brood amount of the target cherax quadricarinatus. According to the method, the accuracy and efficiency of estimating the brood amount of cherax quadricarinatus can be effectively improved, the management cost can be saved while resources are reasonably utilized, and the management efficiency is improved.

Description

Technical Field

[0001] The invention relates to the technical field of aquaculture management, and in particular to a method and a system for improving hatching management efficiency of brooding shrimp. Background Art

[0002] The egg carrying capacity of the parents is a key indicator to measure the reproductive performance of red claw crayfish. In the early stage of red claw crayfish breeding, shrimps with different egg carrying capacities have different requirements for nutrients such as oxygen, trace elements and protein. Red claw crayfish can hatch and produce shrimplets after about 6-7 weeks of carrying eggs. In this process, if the differences in the nutritional needs of the parents are ignored, the management efficiency during the hatching of the seedlings will be low. For example, it is impossible to increase or decrease the amount of feed for individuals with different egg carrying capacities, and blind feeding will also cause a certain amount of feed waste. Therefore, it is necessary to measure the egg carrying capacity of the broodstock during the breeding and seedling raising process, and centrally manage the broodstock with smaller differences in egg carrying capacity. While making rational use of resources, it can also save management costs and improve management efficiency. However, the accuracy and efficiency of the existing methods for measuring the egg carrying capacity of broodstock are low, which affects the management efficiency. Summary of the invention

[0003] The invention overcomes the shortcomings of the prior art and provides a method and a system for improving the hatching management efficiency of brooding shrimp.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses a method for improving hatching management efficiency of brooding shrimp, comprising the following steps:

[0006] Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles;

[0007] Perform feature extraction and denoising on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; perform registration on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model;

[0008] Constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish;

[0009] A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, thereby obtaining the egg carrying capacity of the target red claw crayfish;

[0010] The target red claw crayfish are classified, screened and processed according to their egg carrying capacity, so that parents with different egg carrying capacity can be raised separately.

[0011] More specifically, regional images of the target area are acquired at multiple angles, and image segmentation processing is performed on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles, specifically:

[0012] Acquire a regional image of the target area at multiple angles, convert the regional image into a grayscale image, extract each pixel in the grayscale image and convert it into feature information, so as to represent each pixel in the grayscale image as a feature vector;

[0013] Initialize several cluster centers randomly, calculate the Euclidean distance between each pixel and each cluster center in the grayscale image, and assign it to the cluster with the cluster center with the closest Euclidean distance;

[0014] The feature average of all pixels in each cluster is calculated based on the feature vector of each pixel in each cluster, and is used as the new cluster center;

[0015] Repeat the above steps until the cluster center of each cluster no longer changes, and output the clustering results;

[0016] According to the clustering results, the pixels belonging to the background cluster are marked as background, and the pixels of other clusters are marked as foreground. A binary image is generated based on the marked pixels; the background is black and the foreground is white;

[0017] The binary image is segmented based on a watershed algorithm to remove the background of the regional image and obtain a morphological feature image of the target red claw crayfish.

[0018] More specifically, feature extraction and denoising are performed on each morphological feature image to obtain a feature point set corresponding to each morphological feature image, specifically:

[0019] The SURF feature point detection algorithm is introduced to perform feature extraction processing on each morphological feature image to obtain several feature points of each morphological feature image;

[0020] According to a number of feature points of each morphological feature image, a set of unvisited feature points is created respectively, and each feature point is marked as an unvisited state;

[0021] The DBSCAN algorithm is introduced, and the neighborhood radius and the minimum number of neighbors are preset. The feature points in each feature point set are visited one by one, and the number of neighbors of each visited feature point within the preset radius is counted;

[0022] Compare the number of neighbors of each visited feature point within a preset radius with the minimum number of neighbors;

[0023] If the number of neighbors of a certain visited feature point within the preset radius is greater than or equal to the minimum number of neighbors, the visited feature point is marked as a key feature point;

[0024] If the number of neighbors of a certain visited feature point within the preset radius is less than the minimum number of neighbors, the visited feature point is marked as a noise feature point;

[0025] Repeat the above steps until all feature points in each feature point set are visited; filter out the feature points marked as noise feature points in each feature point set to obtain the feature point set corresponding to each morphological feature image.

[0026] More specifically, the feature point sets corresponding to each morphological feature image are registered to obtain a feature point three-dimensional model, specifically:

[0027] Obtain a feature descriptor for each feature point in each feature point set, and calculate the cosine similarity between any two feature points in each feature point set based on the feature descriptor of each feature point;

[0028] Mark two feature points whose cosine similarity is greater than a preset similarity threshold as a matching feature point pair, and so on, to obtain several pairs of matching feature point pairs;

[0029] Constructing a spatial three-dimensional coordinate system, importing each feature point set into the spatial three-dimensional coordinate system, and performing pairing processing on each feature point set based on matching feature point pairs to obtain an initial feature point three-dimensional model;

[0030] Dividing the initial feature point three-dimensional model into a plurality of sub-regions, and counting the total number of feature points existing in each sub-region;

[0031] If the total number of feature points in a sub-region is less than the preset threshold, the sub-region is marked as a low-density region; if the total number of feature points in a sub-region is greater than the preset threshold, the sub-region is marked as a high-density region;

[0032] If a sub-region is marked as a low-density region, then the relative coordinate values ​​of the feature points in the sub-region in the three-dimensional spatial coordinate system are used to obtain the median coordinate points between any two feature points in the sub-region according to the relative coordinate values, and each of the obtained median coordinate points is set as a new feature point of the sub-region to transform the sub-region into a high-density region;

[0033] Repeat the above steps until all areas marked as low-density areas are converted into high-density areas, and obtain the three-dimensional model of the feature points of the target red claw crayfish.

[0034] More specifically, a 3D model diagram of the morphological characteristics of the target red claw crayfish is constructed according to the 3D model of the characteristic points of the target red claw crayfish, specifically:

[0035] Acquire a feature point three-dimensional model of a target red claw crayfish, randomly select a feature point in the feature point three-dimensional model as a coordinate origin, construct a Cartesian coordinate system according to the coordinate origin, and determine the direction and unit of the coordinate axis;

[0036] Acquire the relative position relationship between the feature points in the Cartesian coordinate system, and connect the feature points using line segments according to the relative position relationship between the feature points;

[0037] Acquire the information of the connection lines between the feature points, perform surface fitting processing on the connection lines between the feature points according to the surface fitting algorithm, and generate a surface model of the target red claw crayfish;

[0038] The surface model of the target red claw crayfish is smoothed and patched to obtain a three-dimensional model of the morphological characteristics of the target red claw crayfish.

[0039] More specifically, a knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, and the egg carrying capacity of the target red claw crayfish is obtained, specifically:

[0040] Collecting characteristic image data of red claw crayfish in various egg carrying ranges through a big data network, constructing a knowledge base, and dividing the knowledge base into a plurality of sub-storage spaces; and assigning a label to each sub-storage space based on the various egg carrying ranges;

[0041] Constructing characteristic three-dimensional model images of red claw crayfish in various egg-carrying ranges according to characteristic image data of red claw crayfish in various egg-carrying ranges; storing the characteristic three-dimensional model images of red claw crayfish in various egg-carrying ranges in sub-storage spaces assigned corresponding labels respectively;

[0042] Obtaining a morphological feature three-dimensional model image of a target red claw crayfish, and calculating the overlap between the morphological feature three-dimensional model image and the feature three-dimensional model images in each sub-storage space based on geometric hashing to obtain a plurality of overlap degrees;

[0043] Extracting a maximum degree of overlap from a plurality of degrees of overlap, marking a sub-storage space corresponding to the maximum degree of overlap, and extracting a label of the marked sub-storage space;

[0044] The egg carrying capacity of the target red claw crayfish is estimated according to the label of the marked sub-storage space.

[0045] A second aspect of the present invention discloses a system for improving hatching management efficiency of brooding shrimps. The system for improving hatching management efficiency of brooding shrimps includes a memory and a processor. The memory stores a method program for improving hatching management efficiency of brooding shrimps. When the method program for improving hatching management efficiency of brooding shrimps is executed by the processor, the following steps are implemented:

[0046] Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles;

[0047] Perform feature extraction and denoising on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; perform registration on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model;

[0048] Constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish;

[0049] A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, thereby obtaining the egg carrying capacity of the target red claw crayfish;

[0050] The target red claw crayfish are classified, screened and processed according to their egg carrying capacity, so that parents with different egg carrying capacity can be raised separately.

[0051] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: acquiring regional images of the target area at multiple angles, and performing image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles; performing feature extraction and denoising processing on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; performing registration processing on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model; constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish; constructing a knowledge base in advance, and importing the morphological feature three-dimensional model diagram of the target red claw crayfish into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, and obtain the egg carrying capacity of the target red claw crayfish; classifying and screening the target red claw crayfish according to the egg carrying capacity of the target red claw crayfish, so as to raise parents with different egg carrying capacities separately. The method not only improves the accuracy and efficiency of the estimation of the egg carrying capacity of the red claw crayfish, but also greatly saves the time and energy of manual analysis, and has high practical value and application potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0053] Figure 1A flow chart of an overall method for improving hatchery management efficiency of brooding shrimp;

[0054] Figure 2 A partial flow chart of a method for improving hatchery management efficiency of brooding shrimp;

[0055] Figure 3 The present invention is a system block diagram of a system for improving hatching management efficiency of brooding shrimp. DETAILED DESCRIPTION

[0056] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0057] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0058] like Figure 1 As shown, the first aspect of the present invention discloses a method for improving hatching management efficiency of brooding shrimp, comprising the following steps:

[0059] S102: acquiring regional images of the target area at multiple angles, and performing image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles;

[0060] S104: performing feature extraction and denoising processing on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; performing registration processing on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model;

[0061] S106: constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish;

[0062] S108: constructing a knowledge base in advance, and importing the three-dimensional model diagram of the morphological characteristics of the target red claw crayfish into the knowledge base to perform an estimation process on the egg carrying capacity of the target red claw crayfish, so as to obtain the egg carrying capacity of the target red claw crayfish;

[0063] S110: Classifying, screening and processing the target red claw crayfish according to the egg carrying capacity of the target red claw crayfish, so as to raise parents with different egg carrying capacity separately.

[0064] It should be noted that, first, by acquiring images of the target area at multiple angles and performing image segmentation processing, the morphological feature images of the target red claw crayfish at different angles are obtained. Then, these morphological feature images are subjected to feature extraction and denoising processing to obtain a clear and accurate feature point set. Through registration processing, these feature point sets are integrated into a three-dimensional model, thereby constructing a three-dimensional model diagram of the morphological features of the target red claw crayfish. In order to improve the accuracy of the egg-carrying capacity measurement, a knowledge base is constructed in advance, and the three-dimensional model diagram of the morphological features of the target red claw crayfish is imported into it. Through this knowledge base, the system can estimate the egg-carrying capacity of the target red claw crayfish, thereby quickly estimating the egg-carrying capacity of the target red claw crayfish and obtaining a more accurate value of the egg-carrying capacity. Finally, according to the egg-carrying capacity of the target red claw crayfish, the system can perform classification and screening processing, and separate the parents with different egg-carrying capacities to optimize breeding management and improve breeding efficiency and yield. The whole process demonstrates a complete technical chain from image acquisition, processing, analysis to knowledge application, realizing automated and precise egg-carrying capacity measurement and management. The application of artificial intelligence technology not only improves the efficiency and accuracy of measurement, but also provides scientific decision-making support for the aquaculture industry, helps to realize intelligent aquaculture management, and promotes the sustainable development of the red claw crayfish aquaculture industry.

[0065] More specifically, if Figure 2 As shown, the regional images of the target area are obtained at multiple angles, and the regional images are subjected to image segmentation processing to obtain the morphological feature images of the target red claw crayfish at multiple angles, specifically:

[0066] S202: acquiring a regional image of the target area at multiple angles, converting the regional image into a grayscale image, extracting each pixel in the grayscale image and converting it into feature information, so as to represent each pixel in the grayscale image as a feature vector;

[0067] S204: randomly initialize a number of cluster centers, calculate the Euclidean distance between each pixel in the grayscale image and each cluster center, and assign the pixel to the cluster where the cluster center with the closest Euclidean distance is located;

[0068] S206: Calculate the feature average of all pixels in each cluster according to the feature vector of each pixel in each cluster, and use it as a new cluster center;

[0069] S208: Repeat the above steps until the cluster center of each cluster no longer changes, and output the clustering result;

[0070] S210: marking pixels belonging to the background cluster as background according to the clustering result, marking pixels of other clusters as foreground, and generating a binary image according to the marked pixels; wherein the background is black and the foreground is white;

[0071] S212: Segmenting the binary image based on a watershed algorithm to remove the background of the regional image and obtain a morphological feature image of the target red claw crayfish.

[0072] It should be noted that, first, the regional image is converted into a grayscale image, and each pixel is converted into a feature vector. Then, by randomly initializing the cluster center and calculating the Euclidean distance between each pixel and the cluster center, the pixel is assigned to the cluster where the nearest cluster center is located. Next, a new cluster center is calculated based on the feature vector of the pixel in each cluster, and this process is repeated until the cluster center no longer changes. Finally, the pixel is marked as background or foreground according to the clustering result, a binary image is generated, and the binary image is segmented using the watershed algorithm to remove the background and obtain the morphological feature image of the target red claw crayfish. This method can effectively remove background interference, extract the morphological characteristics of the target red claw crayfish, and provide a basis for subsequent analysis and processing.

[0073] More specifically, feature extraction and denoising are performed on each morphological feature image to obtain a feature point set corresponding to each morphological feature image, specifically:

[0074] The SURF feature point detection algorithm is introduced to perform feature extraction processing on each morphological feature image to obtain several feature points of each morphological feature image;

[0075] According to a number of feature points of each morphological feature image, a set of unvisited feature points is created respectively, and each feature point is marked as an unvisited state;

[0076] The DBSCAN algorithm is introduced, and the neighborhood radius and the minimum number of neighbors are preset. The feature points in each feature point set are visited one by one, and the number of neighbors of each visited feature point within the preset radius is counted;

[0077] Compare the number of neighbors of each visited feature point within a preset radius with the minimum number of neighbors;

[0078] If the number of neighbors of a certain visited feature point within the preset radius is greater than or equal to the minimum number of neighbors, the visited feature point is marked as a key feature point;

[0079] If the number of neighbors of a certain visited feature point within the preset radius is less than the minimum number of neighbors, the visited feature point is marked as a noise feature point;

[0080] Repeat the above steps until all feature points in each feature point set are visited; filter out the feature points marked as noise feature points in each feature point set to obtain the feature point set corresponding to each morphological feature image.

[0081] It should be noted that the image is feature extracted and noise removed based on the SURF feature point detection algorithm and the DBSCAN clustering algorithm. Specifically, the SURF algorithm is introduced to extract key feature points in each morphological feature image, which usually represent significant features in the image, such as edges, corners, etc. In this way, each morphological feature image can be converted into a set of feature points that describe its important structure. Next, for these feature point sets, an unvisited state list is created to ensure that the same feature point will not be visited repeatedly during the processing process, ensuring the efficiency and accuracy of the algorithm. Subsequently, the DBSCAN algorithm is applied to the feature point set, and the feature points are accessed and analyzed by using the preset neighborhood radius and the minimum number of neighbors as parameters. In this process, each feature point is counted for how many neighboring points are within its preset radius, which is the basis for the DBSCAN algorithm to identify clusters. By comparing the number of neighbors of each visited feature point with the minimum number of neighbors, the algorithm can identify those feature points with enough neighboring feature points as "key feature points", otherwise they are marked as "noise feature points". Key feature points are usually located on important structures or features of the image, while noise feature points may be atypical feature points caused by image noise, blurred edges, etc. Finally, by traversing all feature point sets and marking noise feature points, effective screening of feature points of each morphological feature image is achieved. After removing noise feature points, the remaining key feature point set retains the core information of the morphological feature image, providing a high-quality data foundation for subsequent image analysis, recognition, matching and other tasks. This method realizes the efficient extraction of key features and accurate removal of noise in complex images; improves the stability of feature points and reduces the problem of feature instability caused by image changes (such as illumination, angle changes, etc.); provides high-quality feature descriptors for subsequent image processing tasks, thereby improving the robustness of the system.

[0082] More specifically, the feature point sets corresponding to each morphological feature image are registered to obtain a feature point three-dimensional model, specifically:

[0083] Obtain a feature descriptor for each feature point in each feature point set, and calculate the cosine similarity between any two feature points in each feature point set based on the feature descriptor of each feature point;

[0084] Mark two feature points whose cosine similarity is greater than a preset similarity threshold as a matching feature point pair, and so on, to obtain several pairs of matching feature point pairs;

[0085] Constructing a spatial three-dimensional coordinate system, importing each feature point set into the spatial three-dimensional coordinate system, and performing pairing processing on each feature point set based on matching feature point pairs to obtain an initial feature point three-dimensional model;

[0086] Dividing the initial feature point three-dimensional model into a plurality of sub-regions, and counting the total number of feature points existing in each sub-region;

[0087] If the total number of feature points in a sub-region is less than the preset threshold, the sub-region is marked as a low-density region; if the total number of feature points in a sub-region is greater than the preset threshold, the sub-region is marked as a high-density region;

[0088] If a sub-region is marked as a low-density region, then the relative coordinate values ​​of the feature points in the sub-region in the three-dimensional spatial coordinate system are used to obtain the median coordinate points between any two feature points in the sub-region according to the relative coordinate values, and each of the obtained median coordinate points is set as a new feature point of the sub-region to transform the sub-region into a high-density region;

[0089] Repeat the above steps until all areas marked as low-density areas are converted into high-density areas, and obtain the three-dimensional model of the feature points of the target red claw crayfish.

[0090] It should be noted that, first, by obtaining the feature descriptors of the feature points in each feature point set and calculating the cosine similarity between any two feature points, the feature points with a similarity greater than a preset threshold are marked as matching feature point pairs. Then, each feature point set is imported into the spatial three-dimensional coordinate system, and pairing processing is performed based on the matching feature point pairs to obtain an initial feature point three-dimensional model. Next, the initial feature point three-dimensional model is divided into several sub-regions, and the total number of feature points in each sub-region is counted. According to the comparison between the total number of feature points and the preset threshold, the sub-region is marked as a low-density region or a high-density region. For the low-density region, the relative coordinate values ​​and the median coordinate point of the feature point are calculated, and the median coordinate point is set as a new feature point, thereby converting the low-density region into a high-density region. Repeat this process until all low-density regions are converted into high-density regions, and finally obtain the feature point three-dimensional model of the target red claw crayfish. Through feature point matching and density analysis, a three-dimensional model can be constructed more accurately and the influence of noise and false matching can be reduced. By converting a low-density region into a high-density region, the details and accuracy of the model can be increased, and the shape and structure of the target object can be better reflected. Matching and density analysis of feature points can improve the robustness of the model to different viewing angles, illumination and noise. This method can effectively construct a three-dimensional model of the feature points of the target red claw crayfish, improve the accuracy, details and robustness of the model, and effectively improve the measurement accuracy of the egg load of the red claw crayfish.

[0091] More specifically, a 3D model diagram of the morphological characteristics of the target red claw crayfish is constructed according to the 3D model of the characteristic points of the target red claw crayfish, specifically:

[0092] Acquire a feature point three-dimensional model of a target red claw crayfish, randomly select a feature point in the feature point three-dimensional model as a coordinate origin, construct a Cartesian coordinate system according to the coordinate origin, and determine the direction and unit of the coordinate axis;

[0093] Acquire the relative position relationship between the feature points in the Cartesian coordinate system, and connect the feature points using line segments according to the relative position relationship between the feature points;

[0094] Acquire the information of the connection lines between the feature points, perform surface fitting processing on the connection lines between the feature points according to the surface fitting algorithm, and generate a surface model of the target red claw crayfish;

[0095] The surface model of the target red claw crayfish is smoothed and patched to obtain a three-dimensional model of the morphological characteristics of the target red claw crayfish.

[0096] It should be noted that a feature point is randomly selected from the three-dimensional model of the feature points of the target red claw crayfish as the coordinate origin, and a Cartesian coordinate system is established to determine the direction and unit of the coordinate axis. Then, the relative position relationship between the feature points is obtained in the Cartesian coordinate system, and they are connected using line segments. Next, the information of the connecting line is obtained, and the connecting line is subjected to surface fitting processing using a surface fitting algorithm to generate a surface model of the target red claw crayfish. Finally, the surface model is smoothed and patched to obtain the final morphological feature three-dimensional model diagram. By establishing a Cartesian coordinate system and connecting feature points, the morphological features of the target red claw crayfish, including its shape, size and structure, can be accurately represented. The surface fitting algorithm can generate a smooth surface model based on the information of the connecting line, thereby improving the accuracy and realism of the model. The method can effectively construct a three-dimensional model diagram of the morphological features of the target red claw crayfish.

[0097] More specifically, a knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, and the egg carrying capacity of the target red claw crayfish is obtained, specifically:

[0098] Collecting characteristic image data of red claw crayfish in various egg carrying ranges through a big data network, constructing a knowledge base, and dividing the knowledge base into a plurality of sub-storage spaces; and assigning a label to each sub-storage space based on the various egg carrying ranges;

[0099] Constructing characteristic three-dimensional model images of red claw crayfish in various egg-carrying ranges according to characteristic image data of red claw crayfish in various egg-carrying ranges; storing the characteristic three-dimensional model images of red claw crayfish in various egg-carrying ranges in sub-storage spaces assigned corresponding labels respectively;

[0100] Obtaining a morphological feature three-dimensional model image of a target red claw crayfish, and calculating the overlap between the morphological feature three-dimensional model image and the feature three-dimensional model images in each sub-storage space based on geometric hashing to obtain a plurality of overlap degrees;

[0101] Extracting a maximum degree of overlap from a plurality of degrees of overlap, marking a sub-storage space corresponding to the maximum degree of overlap, and extracting a label of the marked sub-storage space;

[0102] The egg carrying capacity of the target red claw crayfish is estimated according to the label of the marked sub-storage space.

[0103] It should be noted that, first, a large amount of characteristic image data within the egg-carrying capacity range of red claw crayfish was collected through the big data network, and a knowledge base was constructed. This knowledge base is divided into several sub-storage spaces, each of which is marked with a specific egg-carrying capacity range. The purpose of this is to facilitate subsequent retrieval and analysis, so that the model can quickly locate the egg-carrying capacity range closest to the target red claw crayfish. Next, the characteristic three-dimensional model diagrams of red claw crayfish with various egg-carrying capacity ranges are constructed using the characteristic image data, and these model diagrams are stored in the corresponding sub-storage spaces according to their corresponding egg-carrying capacity ranges. This organization method enables the model to efficiently access the egg-carrying capacity model that is most similar to the target red claw crayfish. When it is necessary to measure and estimate the egg-carrying capacity of the target red claw crayfish, the morphological characteristic three-dimensional model diagram of the target red claw crayfish is first obtained. Then, the geometric hash algorithm is used to calculate the overlap between the target model and the models in each sub-storage space, so as to find the egg-carrying capacity range that best matches the target model. The higher the overlap, the more similar the target model is to the model of the egg-carrying capacity range. Among all the calculated overlaps, the largest overlap is selected and the corresponding sub-storage space is marked. The label of the sub-storage space is further extracted, that is, the egg carrying capacity range. Finally, based on this egg carrying capacity range, the approximate egg carrying capacity of the target red claw crayfish can be estimated. By building a knowledge base and using machine learning technology, efficient management and rapid retrieval of a large amount of red claw crayfish image data can be achieved. By calculating the overlap, the egg carrying capacity model that is most similar to the target red claw crayfish can be accurately found, so as to make an accurate egg carrying capacity estimation. This method not only improves the accuracy and efficiency of the red claw crayfish egg carrying capacity estimation, but also greatly saves the time and energy of manual analysis. While making rational use of resources, it can also save management costs and improve management efficiency. It has high practical value and application potential.

[0104] In addition, the method further comprises the following steps:

[0105] Obtaining preset hatching schemes for parents with different egg loads, and determining various preset hatching parameters of each parent hatching equipment when hatching parents with different egg loads according to the preset hatching schemes; the hatching parameters include water temperature, salinity, pH value, and oxygen concentration;

[0106] Acquire various real-time incubation parameters of each parent incubation device, calculate the difference between the various real-time incubation parameters of each parent incubation device and the corresponding preset incubation parameters, and obtain the deviation value of the various real-time incubation parameters of each parent incubation device;

[0107] If the deviation value of a certain real-time incubation parameter of a certain parent incubation device is greater than a preset deviation value, the parent incubation device is marked as an abnormal parent incubation device, and the real-time incubation parameter of the abnormal parent incubation device with a deviation value greater than the preset deviation value is obtained;

[0108] Acquire hatching function characteristic information of each sub-hatching module in the abnormal parent hatching device, and analyze the correlation between the real-time hatching parameters of the abnormal parent hatching device with a deviation value greater than a preset deviation value and the hatching function characteristic information of each sub-hatching module in the abnormal parent hatching device based on the grey correlation analysis method;

[0109] Obtaining a sub-hatching module whose correlation is greater than a preset correlation, which is defined as a correlated sub-hatching module; obtaining real-time working parameters of the correlated sub-hatching module within a preset time period; the working parameters include current, voltage and temperature;

[0110] The real-time working parameters of the relevant sub-incubation module within a preset time period are imported into the support vector machine model for fault diagnosis analysis to obtain the fault probability of the relevant sub-incubation module. If the fault probability is greater than a preset threshold, the cause of the fault is obtained and sent to the user end.

[0111] It should be noted that the incubation parameters of parents with different egg carrying capacity are first determined, and then the deviation value between the real-time incubation parameters and the preset parameters is calculated, the abnormal equipment is marked and the real-time incubation parameters with larger deviation values ​​are obtained. Then, the correlation between the abnormal parameters and the sub-incubation modules is analyzed by the grey correlation analysis method, and the sub-incubation modules with larger correlation and their working parameters are obtained. Finally, the support vector machine model is used to perform fault diagnosis on the relevant sub-incubation modules. If the fault probability is greater than the preset threshold, the cause of the fault is obtained and sent to the user end. This method can realize real-time monitoring and fault diagnosis of parent incubation equipment, improve the reliability and stability of the incubation equipment, reduce the probability of fault occurrence, and thus improve the incubation efficiency and incubation quality. At the same time, the method can also help users to find and solve fault problems in a timely manner, reduce maintenance costs and downtime, and improve management efficiency.

[0112] like Figure 3As shown, the second aspect of the present invention discloses a system for improving hatching management efficiency of brooding shrimps, the system for improving hatching management efficiency of brooding shrimps comprises a memory 41 and a processor 62, the memory 41 stores a method program for improving hatching management efficiency of brooding shrimps, when the method program for improving hatching management efficiency of brooding shrimps is executed by the processor 62, the following steps are implemented:

[0113] Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles;

[0114] Perform feature extraction and denoising on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; perform registration on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model;

[0115] Constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish;

[0116] A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, thereby obtaining the egg carrying capacity of the target red claw crayfish;

[0117] The target red claw crayfish are classified, screened and processed according to their egg carrying capacity, so that parents with different egg carrying capacity can be raised separately.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0119] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0120] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0121] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0122] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0123] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for improving the hatching management efficiency of brooding shrimp, characterized in that: The following steps are involved: Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles; Perform feature extraction and denoising on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; perform registration on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model; Constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish; A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, thereby obtaining the egg carrying capacity of the target red claw crayfish; The target red claw crayfish are classified, screened and processed according to their egg carrying capacity, so that parents with different egg carrying capacity can be raised separately.

2. A method for improving hatching management efficiency of brooding shrimp according to claim 1, characterized in that: Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles, specifically: Acquire a regional image of the target area at multiple angles, convert the regional image into a grayscale image, extract each pixel in the grayscale image and convert it into feature information, so as to represent each pixel in the grayscale image as a feature vector; Initialize several cluster centers randomly, calculate the Euclidean distance between each pixel and each cluster center in the grayscale image, and assign it to the cluster with the cluster center with the closest Euclidean distance; The feature average of all pixels in each cluster is calculated based on the feature vector of each pixel in each cluster, and is used as the new cluster center; Repeat the above steps until the cluster center of each cluster no longer changes, and output the clustering results; According to the clustering results, the pixels belonging to the background cluster are marked as background, and the pixels of other clusters are marked as foreground. A binary image is generated based on the marked pixels; the background is black and the foreground is white; The binary image is segmented based on a watershed algorithm to remove the background of the regional image and obtain a morphological feature image of the target red claw crayfish.

3. A method for improving hatching management efficiency of brooding shrimp according to claim 1, characterized in that: Perform feature extraction and denoising on each morphological feature image to obtain the feature point set corresponding to each morphological feature image, specifically: The SURF feature point detection algorithm is introduced to perform feature extraction processing on each morphological feature image to obtain several feature points of each morphological feature image; According to a number of feature points of each morphological feature image, a set of unvisited feature points is created respectively, and each feature point is marked as an unvisited state; The DBSCAN algorithm is introduced, and the neighborhood radius and the minimum number of neighbors are preset. The feature points in each feature point set are visited one by one, and the number of neighbors of each visited feature point within the preset radius is counted; Compare the number of neighbors of each visited feature point within a preset radius with the minimum number of neighbors; If the number of neighbors of a certain visited feature point within the preset radius is greater than or equal to the minimum number of neighbors, the visited feature point is marked as a key feature point; If the number of neighbors of a certain visited feature point within the preset radius is less than the minimum number of neighbors, the visited feature point is marked as a noise feature point; Repeat the above steps until all feature points in each feature point set are visited; filter out the feature points marked as noise feature points in each feature point set to obtain the feature point set corresponding to each morphological feature image.

4. A method for improving hatching management efficiency of brooding shrimp according to claim 1, characterized in that: The feature point set corresponding to each morphological feature image is registered to obtain a feature point three-dimensional model, which is as follows: Obtain a feature descriptor for each feature point in each feature point set, and calculate the cosine similarity between any two feature points in each feature point set based on the feature descriptor of each feature point; Mark two feature points whose cosine similarity is greater than a preset similarity threshold as a matching feature point pair, and so on, to obtain several pairs of matching feature point pairs; Constructing a spatial three-dimensional coordinate system, importing each feature point set into the spatial three-dimensional coordinate system, and performing pairing processing on each feature point set based on matching feature point pairs to obtain an initial feature point three-dimensional model; Dividing the initial feature point three-dimensional model into a plurality of sub-regions, and counting the total number of feature points existing in each sub-region; If the total number of feature points in a sub-region is less than the preset threshold, the sub-region is marked as a low-density region; if the total number of feature points in a sub-region is greater than the preset threshold, the sub-region is marked as a high-density region; If a sub-region is marked as a low-density region, then the relative coordinate values ​​of the feature points in the sub-region in the three-dimensional spatial coordinate system are used to obtain the median coordinate points between any two feature points in the sub-region according to the relative coordinate values, and each of the obtained median coordinate points is set as a new feature point of the sub-region to transform the sub-region into a high-density region; Repeat the above steps until all areas marked as low-density areas are converted into high-density areas, and obtain the three-dimensional model of the feature points of the target red claw crayfish.

5. A method for improving hatching management efficiency of brooding shrimp according to claim 1, characterized in that: A morphological feature three-dimensional model diagram of the target red claw crayfish is constructed according to the feature point three-dimensional model of the target red claw crayfish, specifically: Acquire a feature point three-dimensional model of a target red claw crayfish, randomly select a feature point in the feature point three-dimensional model as a coordinate origin, construct a Cartesian coordinate system according to the coordinate origin, and determine the direction and unit of the coordinate axis; Acquire the relative position relationship between the feature points in the Cartesian coordinate system, and connect the feature points using line segments according to the relative position relationship between the feature points; Acquire the information of the connection lines between the feature points, perform surface fitting processing on the connection lines between the feature points according to the surface fitting algorithm, and generate a surface model of the target red claw crayfish; The surface model of the target red claw crayfish is smoothed and patched to obtain a three-dimensional model of the morphological characteristics of the target red claw crayfish.

6. A method for improving hatching management efficiency of brooding shrimp according to claim 1, characterized in that: A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, and the egg carrying capacity of the target red claw crayfish is obtained, specifically: Collecting characteristic image data of red claw crayfish in various egg carrying ranges through a big data network, constructing a knowledge base, and dividing the knowledge base into a plurality of sub-storage spaces; and assigning a label to each sub-storage space based on the various egg carrying ranges; According to the characteristic image data of red claw crayfish in various egg carrying ranges, characteristic three-dimensional model diagrams of red claw crayfish in various egg carrying ranges are constructed; The characteristic three-dimensional model images of red claw crayfish with various egg carrying capacity ranges are stored in the sub-storage spaces assigned with corresponding labels; Acquire a morphological feature three-dimensional model image of a target red claw crayfish, and calculate the overlap between the morphological feature three-dimensional model image and the feature three-dimensional model images in each sub-storage space based on geometric hashing to obtain a plurality of overlap degrees; Extracting a maximum degree of overlap from a plurality of degrees of overlap, marking a sub-storage space corresponding to the maximum degree of overlap, and extracting a label of the marked sub-storage space; The egg carrying capacity of the target red claw crayfish is estimated according to the label of the marked sub-storage space.

7. A system for improving hatching management efficiency of brooding shrimp, characterized in that: The system for improving hatching management efficiency of brooding shrimp comprises a memory and a processor, wherein a method program for improving hatching management efficiency of brooding shrimp is stored in the memory, and when the method program for improving hatching management efficiency of brooding shrimp is executed by the processor, the following steps are implemented: Acquire regional images of the target area at multiple angles, and perform image segmentation processing on the regional images to obtain morphological feature images of the target red claw crayfish at multiple angles; Perform feature extraction and denoising on each morphological feature image to obtain a feature point set corresponding to each morphological feature image; perform registration on the feature point set corresponding to each morphological feature image to obtain a feature point three-dimensional model; Constructing a morphological feature three-dimensional model diagram of the target red claw crayfish according to the feature point three-dimensional model of the target red claw crayfish; A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, thereby obtaining the egg carrying capacity of the target red claw crayfish; The target red claw crayfish are classified, screened and processed according to their egg carrying capacity, so that parents with different egg carrying capacity can be raised separately.

8. A system for improving hatching management efficiency of brooding shrimp according to claim 7, characterized in that: The feature point set corresponding to each morphological feature image is registered to obtain a feature point three-dimensional model, which is as follows: Obtain a feature descriptor for each feature point in each feature point set, and calculate the cosine similarity between any two feature points in each feature point set based on the feature descriptor of each feature point; Mark two feature points whose cosine similarity is greater than a preset similarity threshold as a matching feature point pair, and so on, to obtain several pairs of matching feature point pairs; Constructing a spatial three-dimensional coordinate system, importing each feature point set into the spatial three-dimensional coordinate system, and performing pairing processing on each feature point set based on matching feature point pairs to obtain an initial feature point three-dimensional model; Dividing the initial feature point three-dimensional model into a plurality of sub-regions, and counting the total number of feature points existing in each sub-region; If the total number of feature points in a sub-region is less than the preset threshold, the sub-region is marked as a low-density region; if the total number of feature points in a sub-region is greater than the preset threshold, the sub-region is marked as a high-density region; If a sub-region is marked as a low-density region, then the relative coordinate values ​​of the feature points in the sub-region in the three-dimensional spatial coordinate system are used to obtain the median coordinate points between any two feature points in the sub-region according to the relative coordinate values, and each of the obtained median coordinate points is set as a new feature point of the sub-region to transform the sub-region into a high-density region; Repeat the above steps until all areas marked as low-density areas are converted into high-density areas, and obtain the three-dimensional model of the feature points of the target red claw crayfish.

9. A system for improving hatching management efficiency of brooding shrimp according to claim 7, characterized in that: A morphological feature three-dimensional model diagram of the target red claw crayfish is constructed according to the feature point three-dimensional model of the target red claw crayfish, specifically: Acquire a feature point three-dimensional model of a target red claw crayfish, randomly select a feature point in the feature point three-dimensional model as a coordinate origin, construct a Cartesian coordinate system according to the coordinate origin, and determine the direction and unit of the coordinate axis; Acquire the relative position relationship between the feature points in the Cartesian coordinate system, and connect the feature points using line segments according to the relative position relationship between the feature points; Acquire the information of the connection lines between the feature points, perform surface fitting processing on the connection lines between the feature points according to the surface fitting algorithm, and generate a surface model of the target red claw crayfish; The surface model of the target red claw crayfish is smoothed and patched to obtain a three-dimensional model of the morphological characteristics of the target red claw crayfish.

10. The system for improving hatching management efficiency of brooding shrimp according to claim 7, characterized in that: A knowledge base is constructed in advance, and a three-dimensional model diagram of the morphological characteristics of the target red claw crayfish is imported into the knowledge base to estimate the egg carrying capacity of the target red claw crayfish, and the egg carrying capacity of the target red claw crayfish is obtained, specifically: Collecting characteristic image data of red claw crayfish in various egg carrying ranges through a big data network, constructing a knowledge base, and dividing the knowledge base into a plurality of sub-storage spaces; and assigning a label to each sub-storage space based on the various egg carrying ranges; According to the characteristic image data of red claw crayfish in various egg carrying ranges, characteristic three-dimensional model diagrams of red claw crayfish in various egg carrying ranges are constructed; The characteristic three-dimensional model images of red claw crayfish with various egg carrying capacity ranges are stored in the sub-storage spaces assigned with corresponding labels; Acquire a morphological feature three-dimensional model image of a target red claw crayfish, and calculate the overlap between the morphological feature three-dimensional model image and the feature three-dimensional model images in each sub-storage space based on geometric hashing to obtain a plurality of overlap degrees; Extracting a maximum degree of overlap from a plurality of degrees of overlap, marking a sub-storage space corresponding to the maximum degree of overlap, and extracting a label of the marked sub-storage space; The egg carrying capacity of the target red claw crayfish is estimated according to the label of the marked sub-storage space.

Images