Cow and dairy cattle hybridization evaluation system with historical data comparison function

Abstract

The invention discloses a dairy cow and dairy cattle hybridization evaluation system with a historical data comparison function, and relates to the technical field of vertebrate feeding. The method comprises a model construction module, a data acquisition module, a data analysis module and a hybridization evaluation module. By establishing the hybridization evaluation model, the development potential of the calf can be accurately predicted in the young stage of the calf, information reference is provided for follow-up cultivation of the calf, then targeted suggestions can be given based on the individual condition of the calf, the development potential of the calf is matched, the body ability of the calf is fully exerted, and in addition, the development potential of the calf can be accurately predicted. And the model can be optimized and adjusted through body data acquisition of the calf at each stage, so that the prediction accuracy is improved, and the problem that the body ability of the calf is judged only after the development potential of the calf needs to be changed to a great extent for long-time feeding in the existing breeding base is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of vertebrate breeding, and specifically to a hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with a historical data comparison function. Background Art

[0002] Dairy cows, such as Holstein cows, Jersey cows, Kyrgyz cows, etc., these breeds are specially bred for high milk production. Beef and dairy crossbred cattle, such as Simmental cows, Charolais cows, etc., these breeds are suitable for both milk production and meat production. However, in the direction of market optimization, crossing dairy cows with beef and dairy crossbred cattle to obtain better dairy cows or beef and dairy crossbred cattle has always been a popular biological research direction;

[0003] However, due to the viviparous characteristics of cows, one calf is born each year, and usually only one calf per pregnancy. Therefore, the analysis samples of the hybridization results are few and the cycle is long, and the research on the inheritance of excellent genes is slow. For conventional hybridization programs, after obtaining calves in the breeding base, all calves need to be cultivated for a long time to determine their accurate cultivation direction. For example, it can only be judged whether they are suitable for milk production after reproduction, and the meat production situation can only be determined after adulthood. It can only be determined whether they can be used as breeding cattle after giving birth to more offspring or undergoing genetic testing. However, long-term large-scale breeding and high genetic testing costs will result in high breeding costs, and calves with no breeding value cannot be quickly screened out, resulting in limited net profit in the breeding base. In view of this, a hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with a historical data comparison function is provided. Summary of the Invention

[0004] The purpose of the present invention is to provide a hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with a historical data comparison function to solve the existing problems.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: A hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with a historical data comparison function, including:

[0006] A model construction module, which is used to construct a hybrid evaluation model for dairy cows and beef and dairy crossbred cattle;

[0007] A data acquisition module, which is used to establish a historical comparison database by collecting the biological characteristic information of original samples;

[0008] The data acquisition module is also used to acquire the biological characteristic information of the sample to be evaluated;

[0009] A data analysis module, which is used to compare the characteristic information of the sample to be evaluated with the historical comparison database, analyze and classify the sample to be evaluated, and determine the subsequent cultivation direction;

[0010] A hybridization evaluation module, which is used to analyze data through a dairy cow and beef cattle crossbreeding evaluation model, combine continuous data collection feedback on the subsequent breeding of the sample to be evaluated, determine the success rate of the hybridization evaluation result, provide suggestions on cultivation tendencies, and give suggestions for subsequent improvement of the dairy cow and beef cattle crossbreeding evaluation model.

[0011] The present invention also provides a dairy cow and beef cattle crossbreeding evaluation method with a historical data comparison function, including the following steps:

[0012] Construct a dairy cow and beef cattle crossbreeding evaluation model;

[0013] Establish a historical comparison database by collecting the biological characteristic information of the original samples;

[0014] Obtain the biological characteristic information of the sample to be evaluated;

[0015] Breeding evaluation data analysis, by comparing the characteristic information of the sample to be evaluated with the historical comparison database, analyzing and classifying the sample to be evaluated, and determining the subsequent cultivation direction;

[0016] Breeding strategy feedback adjustment, by analyzing data through a dairy cow and beef cattle crossbreeding evaluation model, combining continuous data collection feedback on the subsequent breeding of the sample to be evaluated, determining the success rate of the hybridization evaluation result, providing suggestions on cultivation tendencies, and giving suggestions for subsequent improvement of the dairy cow and beef cattle crossbreeding evaluation model.

[0017] Further, the dairy cow and beef cattle crossbreeding evaluation model includes:

[0018] Construct a dairy cow and beef cattle crossbreeding evaluation model through machine learning algorithms;

[0019] Train the dairy cow and beef cattle crossbreeding evaluation model with historical sample data;

[0020] Deploy the trained dairy cow and beef cattle crossbreeding evaluation model.

[0021] Further, the biological characteristic information data of the original samples is used for the dairy cow and beef cattle crossbreeding evaluation model to conduct a basic evaluation of the development potential of the sample to be evaluated, and the biological characteristic information data of the original samples includes the gene dominant trait data and stage development situation data of the original samples;

[0022] After the dairy cow and beef cattle crossbreeding evaluation model imports the biological characteristic information data of the original samples, the model parameters are improved through the Adam optimizer algorithm.

[0023] Further, the biological characteristic information of the sample to be evaluated includes the external shape characteristic data of the sample to be evaluated;

[0024] Obtain the external feature data of the sample to be evaluated. The external feature data should at least collect the body appearance contour of the sample to be evaluated.

[0025] Furthermore, for the analysis of the breeding evaluation data, by comparing the characteristic information of the sample to be evaluated with the historical comparison database, the sample to be evaluated is analyzed and classified, including:

[0026] Preprocess the characteristic information of the sample to be evaluated and extract features, where the features include head features, ventral features, and side features;

[0027] Analyze the head features to obtain the current milk production potential information and meat quality potential information of the sample to be evaluated;

[0028] Analyze the ventral features to obtain the current reproductive potential information of the sample to be evaluated;

[0029] Analyze the side features to obtain the current body type potential of the sample to be evaluated;

[0030] Based on the analysis results, continue to analyze the subsequent cultivation direction of the sample to be evaluated.

[0031] Furthermore, for the feedback adjustment of the breeding strategy, it is used to determine the success rate of the hybridization evaluation result by combining the analysis data of the dairy cow and beef cattle hybrid evaluation model with the continuous data collection feedback of the subsequent breeding of the sample to be evaluated, provide suggestions on the cultivation tendency, and give suggestions on the subsequent improvement of the dairy cow and beef cattle hybrid evaluation model, including:

[0032] Import the external feature data and sound feature data of the sample to be evaluated into the dairy cow and beef cattle hybrid evaluation model to provide suggestions on the cultivation tendency;

[0033] Provide stage breeding suggestions for the subsequent cultivation tendency of the sample to be evaluated through the dairy cow and beef cattle hybrid evaluation model, and combine the stage development data of the sample to be evaluated to determine the success rate of the hybridization evaluation result;

[0034] Based on the changes in the physical state data during the subsequent cultivation of the sample to be evaluated, continuously supplement new sample information to optimize the historical comparison database of the dairy cow and beef cattle hybrid evaluation model.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] By establishing a hybridization evaluation model, it is possible to accurately predict the development potential of calves at the calf stage, provide information reference for the subsequent cultivation of calves. Secondly, targeted suggestions can be given based on the individual situation of calves to match their development potential and give full play to the physical abilities of calves. Moreover, the model can be optimized and adjusted by collecting the physical data of calves at each stage to increase the accuracy of prediction, solving the problem that existing breeding bases need a long time to feed and only judge the physical abilities of calves after realizing their development potential to a large extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The main flowchart of the hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with historical data comparison function provided by an embodiment of the present invention;

[0038] Figure 2 The specific structural schematic diagram of the hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with historical data comparison function provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] The execution subject of the method in this embodiment is a terminal, and the terminal can be a device such as a mobile phone, a tablet computer, a personal digital assistant (PDA) of a palm computer, a notebook or a desktop computer. Of course, it can also be other devices with similar functions, and this embodiment is not limited thereto.

[0041] Please refer to Figures 1 to 2 , the present invention provides a hybrid evaluation system for dairy cows and beef and dairy crossbred cattle with historical data comparison function, including:

[0042] A model construction module 11, which is used to construct a hybrid evaluation model for dairy cows and beef and dairy crossbred cattle;

[0043] A data acquisition module 12, which is used to establish a historical comparison database by collecting the biological characteristic information of original samples;

[0044] The data acquisition module 12 is also used to acquire the biological characteristic information of the samples to be evaluated;

[0045] A data analysis module 13, which is used to analyze and classify the samples to be evaluated by comparing the characteristic information of the samples to be evaluated with the historical comparison database, and determine the development potential situation;

[0046] The hybrid evaluation module 14 is used to analyze the data of the dairy cow and beef cattle hybrid evaluation model, combine the continuous data collection feedback of the subsequent breeding of the sample to be evaluated, determine the matching rate of the development potential situation, and give suggestions for the subsequent improvement of the dairy cow and beef cattle hybrid evaluation model.

[0047] A hybrid evaluation method for dairy cows and beef cattle with historical data comparison function, comprising the following steps:

[0048] Step 100, constructing a hybrid evaluation model for dairy cows and beef cattle;

[0049] Among them, the step 100 includes:

[0050] Step 110, constructing a hybrid evaluation model through a machine learning algorithm;

[0051] Step 120, training the hybrid evaluation model with historical sample data;

[0052] Step 130, deploying the trained hybrid evaluation model.

[0053] Among them, in the step 110, the machine learning algorithm can adopt algorithms such as neural network, decision tree and Bayesian network to construct a hybrid evaluation model;

[0054] Regarding the neural network, it has a powerful non-linear fitting ability and can well handle the biological characteristic information of the original samples collected (such as the complex relationship between the dominant traits of calves and the development potential). For variables such as biological characteristics that involve complex factors such as the comprehensive influence of multiple genes and the situation of trauma loss, the neural network can learn the patterns in a large amount of data, automatically extract features and establish appropriate mapping relationships;

[0055] Regarding the decision tree algorithm, it has the characteristics of being intuitive and easy to understand. It can make conditional judgments according to different features of the biological characteristic information of the original samples, and construct a tree-like decision structure to predict the development potential of the sample to be evaluated. For biological characteristic information with clear classification features such as the comprehensive influence of multiple genes and the situation of trauma loss, the decision tree can handle it well and incorporate it into the decision-making process;

[0056] Regarding the Bayesian network, it can effectively handle uncertainty and probability relationships. In hybrid evaluation, factors such as the biological characteristic information of the original samples (such as the probability relationship between the dominant traits of calves and different development potentials of calves) can be modeled by the Bayesian network. It can update the prediction probability of the development potential of calves according to prior knowledge (such as the probability distribution of calf potential corresponding to different gene dominant traits) and new evidence (the biological characteristic information of the sample to be evaluated);

[0057] Specifically, which algorithm to adopt is selected according to the specific on-site situation and user needs.

[0058] Taking a neural network as an example, step 100 includes:

[0059] Network structure design:

[0060] Determine the number of input layer nodes; for example, according to the biological characteristic information of the samples, such as head size, face length, ear size, skeleton size, muscle thickness, mammary gland shape, etc.;

[0061] Design the hidden layer. The number of layers and nodes of the hidden layer needs to be determined according to specific circumstances; it can be selected through experiments and experience. For example, for a simple hybridization evaluation model, 1 - 2 hidden layers, and the number of nodes in each layer between 10 - 100 may be a starting range for trial;

[0062] Design the output layer, which can be designed according to the predicted development potential parameters. For example, if predicting two parameters, head size and development potential, the output layer can be set with 2 nodes;

[0063] Data preprocessing and training:

[0064] Normalize the biological characteristic information data of the original samples, and map the data to a smaller range, such as [0, 1] or [-1, 1]. For example, for age, it can be normalized by (face length - minimum face length) / (maximum face length - minimum face length), and for weight, it can be processed in a way such as (mammary gland shape - minimum mammary gland shape) / (maximum mammary gland shape - minimum mammary gland shape);

[0065] Collect a large amount of labeled data, that is, the biological characteristic information of the original samples and the characteristic information of the samples to be evaluated (which can be obtained through image acquisition devices), and divide these data into a training set, a validation set, and a test set. Generally, the ratio can be 6:3:1;

[0066] Use the training set data to train the neural network. Adopt the backpropagation algorithm to update the weights of the network. By calculating the gradient of the loss function with respect to the network weights, use the gradient descent algorithm or its variants (such as optimizers like Adam, Adagrad, etc.) to update the weights to minimize the loss function;

[0067] Model evaluation and deployment:

[0068] Use the validation set to monitor the performance of the model to prevent overfitting. If the loss function starts to increase or the accuracy decreases on the validation set, it indicates that overfitting may occur. It can be improved by adjusting the network structure (such as reducing the number of hidden layer nodes, increasing the regularization term, etc.). Finally, use the test set to evaluate the final performance of the model, obtain a hybridization evaluation model with better generalization ability, and conduct deployment, waiting for subsequent applications.

[0069] Step 200: Establish a historical comparison database by collecting the biological characteristic information of the original samples to improve the parameters of the hybridization evaluation model.

[0070] Among them, the biological characteristic information of the original samples is used for the hybridization evaluation model to conduct a basic assessment of the development potential of calves. The biological characteristic information of the original samples includes, but is not limited to, the gene dominant trait data and stage development situation data of the original samples.

[0071] The gene dominant trait data of the original samples includes at least the head shape data, body shape data, mammary gland appearance of female calves, and genital appearance of male calves of the calves. The head shape data of the calves can further include the face length, eye size, ear length, and nose bridge shape, and the body shape data can include the overall body length, skeleton size, etc.

[0072] Specifically, cows with higher milk production performance usually have a larger head shape. Big eyes and a wide field of vision may help improve the adaptability of dairy cows and their ability to observe the surrounding environment, adapting to the intensive farming environment. A large number of nipples with good shapes helps with efficient milking. A large mammary gland volume and good structure indicate high milk production potential. Cows with higher meat potential usually have a smaller head shape. A flatter head and smaller face length may indicate better meat quality. A large muscle volume and uniform distribution, a strong bone structure help support muscle development, a longer body length helps increase meat production, and larger ears may help regulate body temperature and keep cows comfortable under various environmental conditions.

[0073] The stage development situation data can be obtained through equipment measurement or testing, and can be measured in stages according to time periods such as weeks, half - months, or whole months. The development situation includes, but is not limited to, muscle development, weight gain rate, mammary gland development speed, reproductive cycle, etc., and is summarized to form a historical comparison database.

[0074] After the hybridization evaluation model imports the biological characteristic information data of the original samples, the parameters of the historical comparison database are improved through the gradient descent algorithm.

[0075] Specifically, Step 200 further includes:

[0076] Step 210: First, organize the biological characteristic information of the original samples (such as the gene dominant trait data and stage development situation data of the original samples, etc.) into a format suitable for model input. For example, for a neural network model, normalize these data so that their numerical range is in a suitable interval (such as [0, 1] or [-1, 1]) to improve the efficiency and stability of model training.

[0077] Step 220, after the data is input into the hybridization evaluation model, the model will perform forward propagation to obtain the parameters related to the predicted development status (such as the predicted one-month weight, mammary gland maturation time, etc.). Then, the loss function is calculated based on the predicted values and the actual values (the data of the stage development status collected can be directly used as the actual values);

[0078] Step 230, then, the backpropagation algorithm is used to calculate the gradient of the loss function with respect to the model parameters. In a neural network, the backpropagation algorithm will, according to the chain rule, start from the output layer and calculate the error gradient of each layer in turn and propagate it to the previous layers. For each parameter (such as the weight of the j-th neuron in the i-th layer), its corresponding gradient value will be calculated, indicating the degree of influence of this parameter on the loss function;

[0079] After obtaining the gradient, the gradient descent algorithm can be used to update the model parameters. For example, according to the above update formula, the learning rate is usually a small positive number set in advance (such as 0.001, 0.01, etc.). By continuously repeating this process, that is, iterating the input data, calculating the gradient, and updating the parameters multiple times, the parameters of the model will gradually be adjusted to the state that minimizes the loss function, so that the model can more accurately predict the development potential.

[0080] In step 300, obtain the biological characteristic information of the sample to be evaluated;

[0081] Among them, the biological characteristic information of the sample to be evaluated includes the shape feature data of the sample to be evaluated;

[0082] The shape feature data of the sample to be evaluated is the same as the biological characteristic information data of the original sample, and is the collected appearance dominant shape information;

[0083] Obtain the shape feature data of the sample to be evaluated, and the shape feature data needs to collect at least the body appearance contour of the sample to be evaluated;

[0084] It is understandable that the body appearance contour can be obtained through photography, video recording, taking pictures, etc. When taking pictures, it is necessary to try to ensure that the shooting angle is fixed, the image is centered, and cooperate with image processing algorithms, such as neural network algorithms, to process the pictures, such as unifying the grayscale, inverting the color, etc., to ensure the clarity of the photos and the clarity of the dominant trait details, and reduce the errors caused by image brightness, etc.

[0085] In step 400, analyze the breeding evaluation data, compare the characteristic information of the sample to be evaluated with the historical comparison database, analyze and classify the sample to be evaluated, and determine the development potential;

[0086] Among them, image processing algorithms are used to extract features. For example, HOG (Histogram of Oriented Gradients) can be used to extract the shape edge features of images and is widely used in object detection, and can accurately obtain linear shapes; SIFT (Scale-Invariant Feature Transform) is invariant to image scaling and rotation and is suitable for feature point extraction; SURF (Speeded Up Robust Features) is similar to SIFT but is faster in calculation and is invariant to the affine transformation of images. In this embodiment, no specific limitation is made.

[0087] The features include head features, ventral features, and side features. Classifying the features can shorten the time by only referring to a certain group of features during comparison, or different groups of features can be weighted according to their importance. For example, the ratio of head features: ventral features: side features is 6:2:2.

[0088] Specifically, by analyzing the head features, since the head information contains both milk production-related traits and meat quality-related traits, the current milk production potential information and meat quality potential information of the sample to be evaluated can be obtained after comparison. By analyzing the side features, the current body type potential of the sample to be evaluated can be obtained. The body type potential mainly determines whether it is suitable for meat production, and the calves can be divided into milk-producing types and meat-producing types.

[0089] Specifically, by analyzing the ventral features, genital features and mammary gland features, the current reproductive potential information of the sample to be evaluated can be distinguished. The milk-producing calves and meat-producing calves can be further subdivided into milk-only cows, meat-only beef cattle, milk-producing reproductive cows, meat-producing reproductive cows, milk-producing breeding cows, and meat-producing breeding cows, etc., to optimize the breeding direction and maximize the market value of the calves.

[0090] Combining the analysis results, according to market needs and selling cycles, the subsequent development potential of the sample to be evaluated can be further analyzed to determine a targeted breeding plan. For example, for a breeding base with many dairy factories as customers, only milk-only cows, milk-producing breeding cows, and milk-producing breeding cows can be specifically distinguished, and other types are not specifically distinguished and sold uniformly to reduce the overall cost.

[0091] In step 500, the breeding strategy is feedback-adjusted. Through the analysis data of the dairy cow and milk beef crossbreeding evaluation model, combined with the continuous data collection feedback of the subsequent breeding of the sample to be evaluated, the matching rate of the development potential situation is determined, and suggestions for subsequent improvement of the dairy cow and milk beef crossbreeding evaluation model are given:

[0092] Among them, by importing the appearance feature data of the sample to be evaluated into the dairy cow and dairy-beef cattle hybrid evaluation model, the development potential of the calf can be predicted, such as whether it is suitable for milk production, meat production, suitable as a breeding cow or suitable as a breeding cow, and provide suggestions for training tendencies;

[0093] Through the dairy cow and dairy-beef cattle hybrid evaluation model, stage breeding suggestions are provided for the subsequent cultivation tendency of the samples to be evaluated, and feed is provided and the feeding cycle is refined in a targeted manner. In the actual cultivation, the development data of the samples to be evaluated are collected in cycles, and the success rate of the hybrid evaluation results is determined by combining the stage development data of the samples to be evaluated. If the judgment deviation is large, the relevant parameters can be adjusted in time;

[0094] The biological characteristic information of the samples to be evaluated and the changes in physical condition data collected during subsequent cultivation can be used as new biological characteristic information data of the original samples to supplement the historical comparison database, continuously add new sample information, optimize the historical comparison database of the dairy cow and dairy-beef cattle hybrid evaluation model, and improve the judgment accuracy.

[0095] Those of ordinary skill in the art will appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, equipment and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0097] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0098] In addition, it should be noted that the combination of the various technical features in this case is not limited to the combination described in the claims of this case or the combination described in the specific embodiments. All technical features described in this case can be freely combined or combined in any way unless there is a contradiction between them.

[0099] It should be noted that the above examples are only specific embodiments of the present invention, and the present invention is obviously not limited to the above examples, and there are many similar variations. All variations directly derived or associated from the contents disclosed by the technicians in this field should fall within the protection scope of the present invention.

[0100] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A dairy cow and dairy-beef cow hybrid evaluation system with historical data comparison function, characterized in that: include: A model building module, which is used to build a hybrid evaluation model for dairy cows and dairy-beef cows; A data acquisition module is used to establish a historical comparison database by collecting biological characteristic information of original samples; The data acquisition module is also used to obtain biological characteristic information of the sample to be evaluated; A data analysis module is used to compare the characteristic information of the sample to be evaluated with the historical comparison database, analyze and classify the sample to be evaluated, and determine the development potential; The hybrid evaluation module is used to determine the matching rate of developmental potential through the analysis data of the hybrid evaluation model of dairy cows and dairy-beef cattle, combined with the continuous data collection feedback of the subsequent breeding of the samples to be evaluated, and to provide subsequent improvement suggestions for the hybrid evaluation model of dairy cows and dairy-beef cattle.

2. A method for evaluating the hybridization of dairy cows and dairy beef cattle with a historical data comparison function, using the dairy cow and dairy beef cattle hybridization evaluation system with a historical data comparison function as claimed in claim 1, characterized in that: The following steps are involved: Constructing a hybrid evaluation model for dairy and beef cattle; By collecting biological characteristic information of original samples, a historical comparison database is established; Obtaining biological characteristic information of the sample to be evaluated; Breeding evaluation data analysis: by comparing the characteristic information of the samples to be evaluated with the historical comparison database, the samples to be evaluated are analyzed and classified to determine the development potential; Feedback and adjustment of breeding strategies: through the analysis data of the evaluation model for hybridization of dairy cows and dairy-beef cattle, combined with the continuous data collection and feedback from the subsequent breeding of the samples to be evaluated, the matching rate of developmental potential is determined, and subsequent improvement suggestions for the evaluation model for hybridization of dairy cows and dairy-beef cattle are given.

3. The method for evaluating the hybridization of dairy cows and dairy-beef cattle with historical data comparison function according to claim 2, characterized in that: The hybrid evaluation model of dairy cows and dairy-beef cows comprises: Constructing a dairy and dairy-beef hybrid evaluation model through machine learning algorithms; The evaluation model for dairy and dairy-beef hybrids was trained using historical sample data; Deploy trained dairy and beef cattle cross evaluation models.

4. The method for evaluating the hybridization of dairy cows and dairy-beef cattle with historical data comparison function according to claim 2, characterized in that: The biological characteristic information data of the original sample is used for the dairy cow and dairy-beef cattle hybrid evaluation model to perform a basic evaluation on the development potential of the sample to be evaluated, and the biological characteristic information data of the original sample includes the gene dominant trait data and stage development status data of the original sample; After the biological characteristic information data of the original samples are imported into the dairy cow and dairy-beef cattle hybrid evaluation model, the model parameters are improved through the Adam optimizer algorithm.

5. The method for evaluating the hybridization of dairy cows and dairy-beef cattle with historical data comparison function according to claim 2, characterized in that: The biological characteristic information of the sample to be evaluated, including the appearance characteristic data of the sample to be evaluated; The appearance feature data of the sample to be evaluated is obtained, and the appearance feature data at least needs to collect the body appearance contour of the sample to be evaluated.

6. The method for evaluating the hybridization of dairy cows and dairy-beef cattle with historical data comparison function according to claim 2, characterized in that: The breeding evaluation data analysis is to compare the characteristic information of the sample to be evaluated with the historical comparison database, analyze and classify the sample to be evaluated, and determine the development potential, including: Preprocessing the characteristic information of the sample to be evaluated and extracting features, wherein the features include head features, abdominal bottom features and side body features; Analyze the head features to obtain current milk production potential information and meat quality potential information of the sample to be evaluated; Analyzing the abdominal bottom characteristics to obtain current reproductive potential information of the sample to be evaluated; Analyze the side profile features to obtain the current body shape potential of the sample to be evaluated; Combined with the analysis results, the subsequent developmental potential of the samples to be evaluated will continue to be analyzed.

7. The method for evaluating the hybridization of dairy cows and dairy-beef cattle with historical data comparison function according to claim 2, characterized in that: The breeding strategy feedback adjustment is used to determine the matching rate of development potential through the analysis data of the dairy cow and dairy beef cattle hybrid evaluation model, combined with the continuous data collection feedback of the subsequent breeding of the samples to be evaluated, and provide subsequent improvement suggestions for the dairy cow and dairy beef cattle hybrid evaluation model, including: Import the appearance feature data of the samples to be evaluated into the hybrid evaluation model of dairy cows and dairy-beef cows to provide suggestions for breeding tendencies; The dairy and dairy-beef hybrid evaluation model provides stage breeding suggestions for the subsequent cultivation tendency of the samples to be evaluated, and determines the success rate of the hybrid evaluation results based on the stage development data of the samples to be evaluated; Based on the changes in the physical condition data of the samples to be evaluated during subsequent cultivation, new sample information is continuously added to optimize the historical comparison database of the evaluation model for dairy cows and dairy-beef cattle hybrids.

Images