Yak individual and body size information recognition method and device

Abstract

The application provides a yak individual and body size information recognition method and device, and relates to the technical field of image processing, and comprises the following steps: segmenting the overall yak point cloud data of a to-be-tested yak through a clustering algorithm to obtain yak horn point cloud data; determining the characteristic curve of the yak horn according to the curvature change of each yak horn data point in the yak horn point cloud data, so as to determine the yak horn length information of the to-be-tested yak and the individual information of the to-be-tested yak according to the characteristic curve; determining the age information of the to-be-tested yak according to the yak horn length information and the overall yak point cloud data, so as to obtain the yak hair information corresponding to the age information; determining the yak body point cloud data of the to-be-tested yak according to the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data; and determining the body size measurement information of the to-be-tested yak according to the yak body point cloud data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a yak individual and body size information identification method and device. BACKGROUND

[0002] Traditional yak body size measurement methods mainly rely on manual measurement. Due to the large size of yaks and their strong wildness, multiple adults need to cooperate with each other, consuming a lot of time and effort. The results of manual measurement are easily affected by the subjective factors of the operator, leading to inconsistency of the measurement results. Direct contact with yaks for a long time can trigger the stress response of yaks, affecting their growth and development, and in severe cases, it can lead to serious illness or even death. Contact measurement increases the risk of disease transmission and parasite transmission between humans and animals.

[0003] Therefore, how to effectively identify yak information has become a problem to be solved in the industry. SUMMARY

[0004] The present application provides a yak individual and body size information identification method and device to solve the problem of how to effectively identify yak information in the prior art.

[0005] The present application provides a yak individual and body size information identification method, which includes the following steps.

[0006] The cluster algorithm is used to segment the yak overall point cloud data of the yak to be measured to obtain yak horn point cloud data;

[0007] According to the curvature change of each yak horn data point in the yak horn point cloud data, the characteristic curve of the yak horn is determined, and according to the characteristic curve, the yak horn length information and the yak individual information of the yak to be measured are determined;

[0008] According to the yak horn length information and the yak overall point cloud data, the age information of the yak to be measured is determined, and the yak hair information corresponding to the age information is obtained;

[0009] According to the yak hair point cloud data corresponding to the yak hair information and the yak overall point cloud data, the yak body point cloud data of the yak to be measured is determined;

[0010] According to the yak body point cloud data, the body size measurement information of the yak to be measured is determined.

[0011] According to the yak individual and body size information identification method provided by the present application, before the step of segmenting the yak overall point cloud data of the yak to be measured by the cluster algorithm to obtain the yak horn point cloud data, the method further includes:

[0012] The initial yak point cloud data corresponding to the color depth map of the to-be-tested yak is denoised to obtain denoised yak point cloud data.

[0013] The denoised yak point cloud data is coarsely configured through a point feature histogram, and is finely configured through an iterative closest point algorithm to obtain overall yak point cloud data.

[0014] According to the present application, a yak individual and body size information recognition method is provided, and the feature curve of the yak horn is determined according to the curvature change of each yak horn data point in the yak horn point cloud data, including:

[0015] The first normal vector of each yak horn data point in the yak horn point cloud data and the second normal vector of the corresponding field data point of the yak horn data point are obtained.

[0016] According to the first normal vector of each yak horn data point and the second normal vector of the corresponding field data point of the yak horn data point, the average normal vector curvature of each yak horn data point and the corresponding yak data point corresponding field data point is calculated.

[0017] The yak horn data point with the average normal vector curvature exceeding the preset threshold is taken as the feature curve yak horn data point, and the feature curve of the yak horn is determined according to each feature curve yak horn data point.

[0018] According to the present application, a yak individual and body size information recognition method is provided, and the age information of the to-be-tested yak is determined according to the yak horn length information and the overall yak point cloud data, including:

[0019] According to the overall yak point cloud data, the body height and length information of the to-be-tested yak with hair is determined.

[0020] According to the body height and length information of the to-be-tested yak with hair, the preset regression coefficient, and the yak horn length information, the yak horn length information of the to-be-tested yak is determined through a multiple linear regression analysis method.

[0021] According to the present application, a yak individual and body size information recognition method is provided, and the determination method of the yak hair information corresponding to the age information includes:

[0022] In the preset database, the yak hair information corresponding to the age information is matched.

[0023] The yak hair information includes at least one of the following: hair thickness information, hair length information.

[0024] According to the method for identifying yak individual and body size information provided in the application, after the step of determining the carcass body size measurement information of the to-be-tested yak according to the yak carcass point cloud data, the method further comprises the steps of:

[0025] According to the age information of the to-be-tested yak, the standard carcass body size measurement information corresponding to the age information is determined.

[0026] According to the comparison between the standard carcass body size measurement information and the carcass body size measurement information of the to-be-tested yak, the growth evaluation information of the to-be-tested yak is determined.

[0027] The application further provides a device for identifying yak individual and body size information, comprising the following modules:

[0028] The segmentation module is configured to segment the overall yak point cloud data of the to-be-tested yak by using a clustering algorithm to obtain yak horn point cloud data.

[0029] The first determination module is configured to determine the characteristic curve of the yak horn according to the curvature variation of each yak horn data point in the yak horn point cloud data, and determine the yak horn length information and the individual information of the to-be-tested yak according to the characteristic curve.

[0030] The second determination module is configured to determine the age information of the to-be-tested yak according to the yak horn length information and the overall yak point cloud data, and obtain the yak hair information corresponding to the age information.

[0031] The third determination module is configured to determine the yak carcass point cloud data of the to-be-tested yak according to the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data.

[0032] The identification module is configured to determine the carcass body size measurement information of the to-be-tested yak according to the yak carcass point cloud data.

[0033] The device for identifying yak individual and body size information provided in the application is further configured to:

[0034] The initial yak point cloud data corresponding to the color depth map of the to-be-tested yak is subjected to denoising processing to obtain denoised yak point cloud data.

[0035] The denoised yak point cloud data is subjected to coarse configuration by using a point feature histogram, and the yak point cloud data is subjected to fine configuration by using an iterative closest point algorithm to obtain overall yak point cloud data.

[0036] The device for identifying yak individual and body size information provided in the application is further configured to:

[0037] acquire a first normal vector of each yak horn data point in the yak horn point cloud data and a second normal vector of a field data point corresponding to the yak horn data point;

[0038] calculate a normal vector curvature average value of each yak horn data point and a field data point corresponding to the yak data point according to the first normal vector of each yak horn data point and the second normal vector of the field data point corresponding to the yak horn data point;

[0039] take the yak horn data point with the normal vector curvature average value exceeding a preset threshold as a feature curve yak horn data point, and determine a feature curve of the yak horn according to each feature curve yak horn data point.

[0040] The yak individual and body size information recognition device provided in the application is also used for:

[0041] determining the body height and length information of the yak to be measured with hair according to the overall point cloud data of the yak;

[0042] determining the yak horn length information of the yak to be measured by a multiple linear regression analysis method according to the body height and length information of the yak to be measured with hair, a preset regression coefficient and the yak horn length information.

[0043] The yak individual and body size information recognition device provided in the application is also used for:

[0044] matching the yak hair information corresponding to the age information in a preset database;

[0045] The yak hair information includes at least one of the following: hair thickness information and hair length information.

[0046] The yak individual and body size information recognition device provided in the application is also used for:

[0047] determining the standard carcass body size measurement information corresponding to the age information according to the age information of the yak to be measured;

[0048] determining the growth evaluation information of the yak to be measured according to the comparison between the standard carcass body size measurement information and the carcass body size measurement information of the yak to be measured.

[0049] The application further provides an electronic device including a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the yak individual and body size information recognition method as described above when executing the computer program.

[0050] The application further provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the yak individual and body size information identification method.

[0051] The application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the yak individual and body size information identification method.

[0052] The application provides the yak individual and body size information identification method and device, the cluster algorithm is used to segment the yak whole point cloud data of the yak to be measured, the yak horn point cloud data is obtained, and then the curvature change of each yak horn data point in the yak horn point cloud data is analyzed, the yak horn length information reflecting the age of the yak is effectively determined, so that the yak hair information corresponding to the age information can be obtained, the yak body point cloud data of the yak to be measured is determined through the yak hair point cloud data corresponding to the yak hair information and the yak whole point cloud data, the existence of the hair can interfere with the measurement result, and therefore more accurate body size measurement information can be obtained. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0054] Figure 1 The application provides a yak individual and body size information identification method flowchart.

[0055] Figure 2 The application provides a variable geometry schematic diagram.

[0056] Figure 3 The application provides a yak individual and body size information identification device structure schematic diagram.

[0057] Figure 4 The application provides an electronic device structure schematic diagram. DETAILED DESCRIPTION

[0058] In order to make the objects, technical solutions and advantages of the application clearer, the following will combine the drawings in the application to clearly and completely describe the technical solutions in the application. Obviously, the described embodiments are some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative effort belong to the protection scope of the application.

[0059] Figure 1 The flowchart of the yak individual and body size information recognition method provided by the present application is shown in FIG. 1, which includes the following steps: Figure 1

[0060] In step 110, the overall point cloud data of the yak to be measured is segmented by a clustering algorithm to obtain yak horn point cloud data.

[0061] In the embodiments of the present application, the clustering algorithm can be K-means or spectral clustering algorithm. The point cloud is segmented by the clustering algorithm to obtain yak horn point cloud data.

[0062] K-means clustering algorithm (K-Means Clustering) is an iterative optimization algorithm based on distance measurement. By selecting several centroids, the data is grouped so that the sum of distances from each data point to the centroid within the cluster is minimized.

[0063] K initial centroids are randomly selected, or a heuristic method such as K-Means++ is used to better initialize the centroids, reducing the impact of randomness on clustering results. Each data point is assigned to the cluster where the nearest centroid is located. This step involves calculating the distance between data points and centroids, and assigning data points to the cluster represented by the nearest centroid.

[0064] The centroid of each cluster is recalculated, i.e. the average value of all data points in the cluster is taken to update the position of the centroid. Repeat the above steps until the position of the centroid no longer changes, or the preset number of iterations is reached, at which point the algorithm converges.

[0065] According to the clustering results, the point cloud data belonging to the yak horn is recognized and extracted. Due to the differences in geometry and structure between the yak horn and other body parts, they will be assigned to different clusters, thereby achieving segmentation.

[0066] The silhouette coefficient (Silhouette Coefficient) and other clustering effectiveness indicators can be used to evaluate the quality of the clustering results. The silhouette coefficient reflects the intra-cluster tightness and inter-cluster separation of the cluster structure. The k value with the maximum average silhouette coefficient is the optimal number of clusters, and the yak horn point cloud data is finally obtained.

[0067] In step 120, the feature curve of the yak horn is determined according to the curvature change of each yak horn data point in the yak horn point cloud data, and the yak horn length information and yak individual information of the yak to be measured are determined according to the feature curve.

[0068] ​In this invention, the local curvature of each data point in the yak horn point cloud is calculated. The curvature can be determined by the relationship between the normal vector of each point in the point cloud and the normal vectors of its neighboring points. Geometric information in the point cloud, such as the local normal and radius of curvature of a point, can be used to calculate the curvature.

[0069] Identify points with significant changes in curvature as feature points. These feature points may correspond to the edges or contours of a yak horn and are key to determining the characteristic curve.

[0070] Based on feature points, the characteristic curve of the yak horn is determined. This can be achieved by connecting points with significant changes in curvature to form a curve representing the outline of the yak horn. The length of the yak horn is then determined by calculating the length of this characteristic curve. This can be done by calculating the distances between consecutive points on the curve and summing them.

[0071] Since each yak's horn is unique, this invention allows for individual yak identification by combining the characteristic curves and length information of the yak horn with data from a yak horn feature database. This step involves feature matching algorithms, such as pattern recognition-based methods, which compare newly acquired yak horn features with features in the database to identify the individual yak information.

[0072] Step 130: Based on the yak horn length information and the overall point cloud data of the yak, determine the age information of the yak to be tested, so as to obtain the yak hair information corresponding to the age information;

[0073] The length of a yak's horns is directly related to its age. As a yak ages, its horns gradually grow longer, forming more whorls. By measuring the length of a yak's horns, its age can be effectively determined.

[0074] Specifically, multiple linear regression analysis can be used to predict yak age by combining yak horn length information and yak body size data. The linear regression model is as follows:

[0075]

[0076] in, It's age. This is information about the length of yak horns. arrive It is the regression coefficient. arrive These are different body size parameters of yaks. It's an error.

[0077] In this invention, based on the predicted age of the yak, information about the yak's hair corresponding to that age is obtained. This information may include characteristics such as hair length and density.

[0078] Step 140, according to the yak hair point cloud data corresponding to the yak hair information and the yak overall point cloud data, determine the yak carcass point cloud data of the yak to be measured;

[0079] In the present application, according to the characteristics of yak hair point cloud data, such as length, density, etc., it is distinguished from yak carcass point cloud data. This can be done through feature calculation and analysis of point cloud data.

[0080] After removing the hair point cloud data from the yak overall point cloud data, the remaining is the yak carcass point cloud data. This step ensures the accuracy and integrity of the carcass point cloud data

[0081] Step 150, according to the yak carcass point cloud data, determine the carcass size measurement information of the yak to be measured.

[0082] In the present application, the key feature points in the yak carcass point cloud data are identified, such as the top point, the bottom edge, the side surface, etc., and the distance between the feature points is measured by using point cloud data processing software or algorithm. These distances include the length, width, height, etc. of the carcass.

[0083] According to the measured distance, the carcass size parameters such as body height, body length, chest circumference, hip circumference, etc. are calculated. These parameters can be calculated by geometric method or statistical method.

[0084] In the present application, by clustering algorithm, the yak overall point cloud data of the yak to be measured is segmented to obtain the yak horn point cloud data, and then by analyzing the curvature change of each yak horn data point in the yak horn point cloud data, the yak horn length information reflecting the age of the yak is effectively determined, so that the age information corresponding to the yak hair information can be obtained, and by the yak hair point cloud data corresponding to the yak hair information and the yak overall point cloud data, the yak carcass point cloud data of the yak to be measured is determined. The existence of hair may interfere with the measurement results, so as to obtain more accurate carcass size measurement information.

[0085] Optionally, before the step of segmenting the yak overall point cloud data of the yak to be measured by clustering algorithm to obtain the yak horn point cloud data, the method further comprises:

[0086] The initial yak point cloud data corresponding to the color depth map of the yak to be measured is denoised to obtain the denoised yak point cloud data;

[0087] The denoised yak point cloud data is coarsely configured by point feature histogram, and the yak point cloud data is finely configured by iterative closest point algorithm to obtain the yak overall point cloud data.

[0088] In the present application, first, the side view and overhead view depth images of the yak are collected by means of a dual RGB-D camera, obtaining the color depth map of the yak to be measured. The initial yak point cloud data of the yak to be measured is obtained through the color depth map. These data contain the three-dimensional geometric information and color information of the yak.

[0089] The initial yak point cloud data is denoised. The noise points and outliers in the point cloud are removed, which may be caused by sensor errors, environmental interference or errors in the data acquisition process.

[0090] The denoised yak point cloud data is coarsely registered by the Point Feature Histogram (PFH) algorithm. PFH is a method for describing the local geometric features of point clouds, which can capture the neighborhood shape features of each point in the point cloud. The purpose of coarse registration is to determine the approximate relative position and pose between two or more point cloud data sets, providing an initial estimate for fine registration.

[0091] The yak point cloud data after coarse registration is fine registered by the Iterative Closest Point (ICP) algorithm. The ICP algorithm optimizes the alignment between point clouds by iteratively finding corresponding point pairs and minimizing the distance between them. Fine registration can improve the alignment accuracy of point cloud data and ensure accurate matching between point cloud data sets.

[0092] After coarse registration and fine registration, the overall yak point cloud data is obtained, which accurately reflects the three-dimensional geometric structure of the yak to be measured, providing high-quality input data for subsequent clustering segmentation and feature extraction.

[0093] Optionally, according to the curvature change of each yak horn data point in the yak horn point cloud data, the characteristic curve of the yak horn is determined, including:

[0094] The first normal vector of each yak horn data point in the yak horn point cloud data and the second normal vector of the corresponding field data point of the yak horn data point are obtained.

[0095] According to the first normal vector of each yak horn data point and the second normal vector of the corresponding field data point of the yak horn data point, the average normal vector curvature of each yak horn data point and the corresponding field data point of the yak data point is calculated.

[0096] The yak horn data point with the average normal vector curvature exceeding the preset threshold is taken as the characteristic curve yak horn data point, and the characteristic curve of the yak horn is determined according to each characteristic curve yak horn data point.

[0097] In the scheme of the present invention, for each data point in the yak horn point cloud data, its first normal vector is calculated. This is usually done by estimating the local surface around the point and calculating its surface normal. A field (neighborhood) is determined for each yak horn data point, which is usually a spherical or cubic region centered at the point. For the neighborhood data points within the field of each yak horn data point, their second normal vectors are calculated.

[0098] Using the first normal vector of each yak horn data point and the second normal vectors of the neighborhood data points, the normal curvature of these points is calculated. Curvature can be estimated by measuring the degree of concave and convex of the local surface.

[0099] For each yak horn data point, the average of its normal curvature with the neighborhood data points is calculated. This average can provide a quantitative measure of the local geometric variation of the point.

[0100] A preset curvature threshold is determined to identify points with significant curvature changes, which are likely to be feature points of the yak horn. Yak horn data points with normal curvature average exceeding the preset threshold are identified as feature curve yak horn data points. These points represent significant geometric features of the yak horn, such as the edges or curved parts of the horn.

[0101] All identified feature curve yak horn data points are connected to form the feature curve of the yak horn.

[0102] More specifically, for each point p in the point cloud, let N be the unit normal vector of point p, and assume that p has m nearest neighbors, the normal vector of the i-th nearest neighbor of point P is Let the orthogonal coordinate system

[0103] be L for point p, where N represents the normal vector of point p. X and Y are orthogonal unit vectors. In L, the coordinates of are , , . The normal curvature of point p is estimated using an osculating circle passing through point p . Figure 2 The variable geometry diagram provided by the present invention is shown in Figure 2 .

[0104] Then the normal curvature of p relative to is estimated as follows:

[0105]

[0106] where,​​ is the angle between vectors and is the angle between vectors and

[0107] The approximation of equation (1.1-1) is as follows:

[0108] (1.1-2)

[0109] where, ,

[0110] The length of the feature curve L is calculated using the Euclidean distance. Assuming that the curve L is composed of n points , , , , , then the arc length L can be approximated as:

[0111]

[0112] In the scheme of the present application, the feature curve representing the geometric features of the yak horn can be accurately extracted from the yak horn point cloud data, which is crucial for subsequent applications such as yak horn length measurement, morphological analysis, and yak individual identification.

[0113] Optionally, the method for determining the yak hair information corresponding to the age information comprises:

[0114] matching the yak hair information corresponding to the age information in a preset database;

[0115] wherein the yak hair information comprises at least one of the following: hair thickness information, hair length information.

[0116] In the present application, in the preset database, the yak hair information corresponding to the age is matched according to the determined age information of the yak. This database may contain hair feature data of yaks of different ages, including hair thickness and length information. By matching the information in the preset database, the hair features of yaks of a certain age can be determined

[0117] Optionally, after the step of determining the carcass size measurement information of the yak to be measured according to the carcass point cloud data of the yak, the method further comprises:

[0118] determining the standard carcass size measurement information corresponding to the age information according to the age information of the yak to be measured;

[0119] ​​​According to the comparison of the standard carcass size measurement information and the carcass size measurement information of the yak to be measured, growth evaluation information of the yak to be measured is determined.

[0120] In the present application, according to the age information of the yak to be measured, the standard carcass size measurement information corresponding to the age is matched in a preset database. This database may contain standard size data of yaks at different age stages, such as body height, body length and chest circumference, etc.

[0121] The carcass size measurement information of the yak to be measured is compared with the standard size information in the database, and according to the comparison result, the growth evaluation information of the yak to be measured is determined. If the size measurement value of the yak to be measured is close to or better than the standard value, it can be considered that the growth and development condition is good; if it is lower than the standard value, further nutrition or health management may be needed.

[0122] In the present application, the growth evaluation information can be used to guide the feeding management of yaks, such as adjusting the feed ratio, improving the feeding environment, etc., to promote the healthy growth of yaks

[0123] In an optional embodiment, a camera, scanner or other equipment is used to collect images or videos of the horn of the yak to obtain comprehensive horn information. By collecting images from multiple angles, the characteristics of the yak horn can be more accurately captured.

[0124] The collected images are preprocessed, including denoising and contrast enhancement, etc., to improve the image quality. Subsequently, the yak horn is separated from the background through segmentation algorithm, laying a foundation for subsequent feature extraction and analysis.

[0125] Key features are extracted from the segmented yak horn image, such as horn length, bending degree and texture, etc. This step can be realized by using morphological operation, edge detection and texture analysis, etc.

[0126] A feature database containing known yak horn feature data is established. Then, the newly collected yak horn features are compared with the features in the database using a feature matching algorithm (such as a pattern recognition based method) to identify the individual information of the yak.

[0127] According to the feature data of the yak horn, the length, bending degree, etc. are calculated. By comparing with the size data of known individuals, the growth and development of the yak can be evaluated.

[0128] In another optional embodiment, data including the size, condition indicators and corresponding hair parameters of the yak are collected. Ensure that the collected data is representative and extensive to cover different breeds, ages and genders.

[0129] The collected data is pre-processed, including data cleaning and outlier processing, to ensure the accuracy and consistency of the data.

[0130] According to the actual needs, the body size and body condition indicators with high correlation with the hair parameters are selected as the features. The features with high correlation can be determined by correlation analysis and other methods.

[0131] According to the problem requirements, appropriate fitting models are selected, such as linear regression, polynomial regression, support vector regression, etc. The selected features and corresponding hair parameters are used as input and output to train the model, and the training set is used to fit and optimize the model parameters.

[0132] The established model is evaluated and verified using the test set. The accuracy and generalization ability of the model are evaluated by calculating the prediction error, root mean square error, etc.

[0133] The verified fitting model can be applied to actual data prediction and analysis. By inputting the body size and body condition indicators of the yak, the corresponding hair parameters can be predicted, providing a scientific basis for the breeding and management of yaks.

[0134] The yak individual and body size information recognition device provided by the present application is described below. The yak individual and body size information recognition device described below can be referred to in conjunction with the yak individual and body size information recognition method described above.

[0135] Figure 3 The structure diagram of the yak individual and body size information recognition device provided by the present application is shown in Figure 3 , which includes:

[0136] The segmentation module 310 is used to segment the overall yak point cloud data of the yak to be tested by a clustering algorithm to obtain yak horn point cloud data.

[0137] The first determination module 320 is used to determine the feature curve of the yak horn according to the curvature change of each yak horn data point in the yak horn point cloud data, so as to determine the yak horn length information and individual information of the yak to be tested according to the feature curve.

[0138] The second determination module 330 is used to determine the age information of the yak to be tested according to the yak horn length information and the overall yak point cloud data, so as to obtain the yak hair information corresponding to the age information.

[0139] The third determination module 340 is used to determine the yak body point cloud data of the yak to be tested according to the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data.

[0140] The recognition module 350 is configured to determine the carcass size measurement information of the to-be-tested yak according to the yak carcass point cloud data.

[0141] The yak individual and size information recognition device provided in the present application is also used for:

[0142] The initial yak point cloud data corresponding to the color depth map of the to-be-tested yak is denoised to obtain denoised yak point cloud data.

[0143] The denoised yak point cloud data is coarsely configured through a point feature histogram, and the yak point cloud data is finely configured through an iterative closest point algorithm to obtain overall yak point cloud data.

[0144] The yak individual and size information recognition device provided in the present application is also used for:

[0145] The first normal vector of each yak horn data point in the yak horn point cloud data and the second normal vector of the field data point corresponding to the yak horn data point are obtained.

[0146] The average value of the normal vector curvature of each yak horn data point and the corresponding field data point of the corresponding yak data point is calculated according to the first normal vector of each yak horn data point and the second normal vector of the field data point corresponding to the yak horn data point.

[0147] The yak horn data point with the average value of the normal vector curvature exceeding a preset threshold is taken as a feature curve yak horn data point, and the feature curve of the yak horn is determined according to each feature curve yak horn data point.

[0148] The yak individual and size information recognition device provided in the present application is also used for:

[0149] The body height and length information of the to-be-tested yak with hair is determined according to the overall yak point cloud data.

[0150] The yak horn length information of the to-be-tested yak is determined according to the body height and length information of the to-be-tested yak with hair, a preset regression coefficient, and the yak horn length information through a multivariate linear regression analysis method.

[0151] The yak individual and size information recognition device provided in the present application is also used for:

[0152] The yak hair information corresponding to the age information is matched in a preset database.

[0153] The yak hair information includes at least one of the following: hair thickness information, hair length information.

[0154] The device is also used for:

[0155] According to the age information of the to-be-tested yak, standard carcass body size measurement information corresponding to the age information is determined.

[0156] According to comparison of the standard carcass body size measurement information and the carcass body size measurement information of the to-be-tested yak, growth evaluation information of the to-be-tested yak is determined.

[0157] In the present application, the overall yak point cloud data of the to-be-tested yak is segmented by a clustering algorithm to obtain yak horn point cloud data, and then the curvature change of each yak horn data point in the yak horn point cloud data is analyzed to effectively determine yak horn length information that can reflect the age of the yak, so that yak hair information corresponding to the age information can be obtained, and the yak carcass point cloud data of the to-be-tested yak is determined by the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data, and the existence of hair may interfere with the measurement result, so that more accurate carcass body size measurement information is obtained.

[0158] Figure 4 is a structural schematic diagram of an electronic device provided by the present application, as Figure 4 shown, the electronic device can include a processor (processor) 410, a communications interface (Communications Interface) 420, a memory (memory) 430 and a communication bus 440, wherein the processor 410, the communication interface 420, the memory 430 complete the communication among each other through the communication bus 440. The processor 410 can call the logic instructions in the memory 430 to execute the yak individual and body size information identification method, which comprises: segmenting the overall yak point cloud data of the to-be-tested yak by a clustering algorithm to obtain yak horn point cloud data;

[0159] According to the curvature change of each yak horn data point in the yak horn point cloud data, a characteristic curve of the yak horn is determined, so as to determine the yak horn length information and the to-be-tested yak individual information of the to-be-tested yak according to the characteristic curve.

[0160] According to the yak horn length information and the overall yak point cloud data, the age information of the to-be-tested yak is determined to obtain yak hair information corresponding to the age information.

[0161] According to the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data, the yak carcass point cloud data of the to-be-tested yak is determined.

[0162] According to the yak carcass point cloud data, the carcass body size measurement information of the to-be-tested yak is determined.

[0163] Moreover, the logic instructions in the memory 430 described above can be realized in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0164] On the other hand, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the yak individual and body size information identification method provided by the above-mentioned method, the method comprising: segmenting the yak overall point cloud data of the to-be-tested yak by a clustering algorithm to obtain yak horn point cloud data;

[0165] According to the curvature change of each yak horn data point in the yak horn point cloud data, a characteristic curve of the yak horn is determined, so as to determine the yak horn length information and the yak individual information of the to-be-tested yak according to the characteristic curve;

[0166] According to the yak horn length information and the yak overall point cloud data, the age information of the to-be-tested yak is determined, so as to obtain yak hair information corresponding to the age information;

[0167] According to the yak hair point cloud data corresponding to the yak hair information and the yak overall point cloud data, yak body point cloud data of the to-be-tested yak is determined;

[0168] According to the yak body point cloud data, the body size measurement information of the to-be-tested yak is determined.

[0169] On the other hand, the present application also provides a non-transitory computer readable storage medium, which stores a computer program, the computer program is executed by a processor to realize the yak individual and body size information identification method provided by the above-mentioned method, the method comprising: segmenting the yak overall point cloud data of the to-be-tested yak by a clustering algorithm to obtain yak horn point cloud data;

[0170] According to the curvature change of each yak horn data point in the yak horn point cloud data, a characteristic curve of the yak horn is determined, and according to the characteristic curve, the yak horn length information of the to-be-tested yak and individual information of the to-be-tested yak are determined;

[0171] According to the yak horn length information and the overall point cloud data of the yak, age information of the to-be-tested yak is determined, so as to obtain yak hair information corresponding to the age information;

[0172] According to the yak hair point cloud data corresponding to the yak hair information and the overall point cloud data of the yak, yak body point cloud data of the to-be-tested yak is determined;

[0173] According to the yak body point cloud data, body size measurement information of the to-be-tested yak is determined.

[0174] The device embodiments described above are only schematic, wherein the units shown as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment scheme. Those skilled in the art can understand and implement without creative labor.

[0175] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software product, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including a plurality of instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute the method described in each embodiment or some part of the embodiment.

[0176] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying individual yaks and their body size information, characterized in that, include: By using a clustering algorithm, the overall point cloud data of the yak to be tested is segmented to obtain the yak horn point cloud data; Based on the curvature changes of each yak horn data point in the yak horn point cloud data, the characteristic curve of the yak horn is determined, and the yak horn length information and individual information of the yak to be tested are determined based on the characteristic curve; Based on the yak horn length information and the overall point cloud data of the yak, the age information of the yak to be tested is determined, so as to obtain the yak hair information corresponding to the age information; Based on the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data, the yak carcass point cloud data of the yak to be tested is determined; Based on the yak carcass point cloud data, the body size measurement information of the yak to be tested is determined; Specifically, the characteristic curve of the yak horn is determined based on the curvature changes of each yak horn data point in the yak horn point cloud data, including: Obtain the first normal vector of each yak horn data point in the yak horn point cloud data and the second normal vector of the corresponding neighborhood data point; Based on the first normal vector of each yak horn data point and the second normal vector of the corresponding neighborhood data point, calculate the average curvature of the normal vectors of each yak horn data point and the corresponding neighborhood data point. The yak horn data points whose average normal vector curvature exceeds a preset threshold are used as characteristic curve yak horn data points, and the characteristic curve of the yak horn is determined based on each characteristic curve yak horn data point. The step of determining the age information of the yak to be tested based on the yak horn length information and the overall point cloud data of the yak includes: Based on the overall point cloud data of the yak, the body height and length information of the yak with hair were determined; Using multiple linear regression analysis, based on the body height and length of the yak with hair, preset regression coefficients, and the yak horn length information, the age information of the yak is determined.

2. The method for identifying individual yaks and their body size information according to claim 1, characterized in that, Before the step of segmenting the overall yak point cloud data of the yak to be tested using a clustering algorithm to obtain yak horn point cloud data, the method further includes: The initial yak point cloud data corresponding to the color depth map of the yak to be tested is denoised to obtain denoised yak point cloud data; The denoised yak point cloud data is coarsely configured using point feature histograms, and finely configured using an iterative nearest-point algorithm to obtain the overall yak point cloud data.

3. The method for identifying individual yaks and their body size information according to claim 1, characterized in that, The method for determining the yak hair information corresponding to the age information includes: In a preset database, match yak hair information corresponding to the age information; The yak hair information includes at least one of the following: hair thickness information and hair length information.

4. The method for identifying individual yaks and their body size information according to claim 1, characterized in that, After determining the body size measurement information of the yak carcass based on the yak carcass point cloud data, the method further includes: Based on the age information of the yak to be tested, determine the standard carcass body size measurement information corresponding to the age information; The growth assessment information of the yak under test is determined by comparing the standard carcass body size measurement information with the carcass body size measurement information of the yak under test.

5. A device for identifying individual yaks and their body size information, characterized in that, include: The segmentation module is used to segment the overall point cloud data of the yak to be tested using a clustering algorithm to obtain the yak horn point cloud data; The first determining module is used to determine the characteristic curve of the yak horn based on the curvature change of each yak horn data point in the yak horn point cloud data, so as to determine the yak horn length information and individual information of the yak to be tested based on the characteristic curve; The second determining module is used to determine the age information of the yak to be tested based on the yak horn length information and the overall point cloud data of the yak, so as to obtain the yak hair information corresponding to the age information; The third determining module is used to determine the yak carcass point cloud data of the yak to be tested based on the yak hair point cloud data corresponding to the yak hair information and the overall yak point cloud data; The identification module is used to determine the body size measurement information of the yak carcass based on the yak carcass point cloud data; Specifically, the characteristic curve of the yak horn is determined based on the curvature changes of each yak horn data point in the yak horn point cloud data, including: Obtain the first normal vector of each yak horn data point in the yak horn point cloud data and the second normal vector of the corresponding neighborhood data point; Based on the first normal vector of each yak horn data point and the second normal vector of the corresponding neighborhood data point, calculate the average curvature of the normal vectors of each yak horn data point and the corresponding neighborhood data point. The yak horn data points whose average normal vector curvature exceeds a preset threshold are used as characteristic curve yak horn data points, and the characteristic curve of the yak horn is determined based on each characteristic curve yak horn data point. The step of determining the age information of the yak to be tested based on the yak horn length information and the overall point cloud data of the yak includes: Based on the overall point cloud data of the yak, the body height and length information of the yak with hair were determined; Using multiple linear regression analysis, based on the body height and length of the yak with hair, preset regression coefficients, and the yak horn length information, the age information of the yak is determined.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for identifying individual yaks and their body size information as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for identifying individual yaks and their body size information as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for identifying individual yaks and their body size information as described in any one of claims 1 to 4.

Images