Method and system for predicting growth performance of three-way hybrid pigs

Abstract

The invention relates to a growth performance prediction method and system for three-way hybrid pigs, and belongs to the technical field of data processing. The method comprises the following steps: acquiring growth characteristic data of N three-way hybrid pigs in a target pigsty and a blood relationship of each three-way hybrid pig; wherein the growth characteristic data comprises body weight, feed intake, backfat thickness, eye muscle area and eye muscle depth; the growth characteristic data are periodically acquired time sequence data; n is a positive integer; constructing a blood relationship map based on the blood relationship of each three-way hybrid pig; and inputting the growth characteristic data of the N three-way hybrid pigs in the target pigsty and the blood relationship graph into a set dynamic graph convolution model, and outputting the growth performance data of each three-way hybrid pig at the next acquisition node. Through the process, the potential influence of the genotype on the growth of the three-way hybrid pig can be better understood, so that the accuracy of predicting the growth performance of the three-way hybrid pig can be improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for predicting the growth performance of three-way crossbred pigs. Background Technology

[0002] Three-way crossbred pigs are offspring produced by crossing three different breeds of pigs to obtain offspring with better production performance, adaptability, and meat quality. Three-way crossbreeding is a commonly used breeding strategy in modern pig breeding to fully utilize the advantages of different breeds and compensate for the shortcomings of a single breed.

[0003] Since three-way crossbred pigs are obtained by crossing three different breeds of pigs, predicting their growth performance is extremely important. Predicting the growth performance of three-way crossbred pigs helps farms understand the growth potential of piglets in advance, select breeding pigs with excellent growth characteristics, and identify potential health problems, among other things.

[0004] However, research has found that existing predictions of pig growth performance are based solely on traditional empirical data and cannot be effectively applied to three-way crossbred pigs, resulting in very low accuracy in predicting the growth performance of three-way crossbred pigs. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, this invention provides a method and system for predicting the growth performance of three-way crossbred pigs.

[0006] In a first aspect, embodiments of this application provide a method for predicting the growth performance of three-way crossbred pigs, comprising: acquiring growth characteristic data of N three-way crossbred pigs in a target pigpen, and the pedigree of each three-way crossbred pig; wherein the growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the growth characteristic data are all periodically collected time-series data; N is a positive integer; constructing a pedigree map based on the pedigree of each three-way crossbred pig; inputting the growth characteristic data of the N three-way crossbred pigs in the target pigpen, and the pedigree map, into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next acquisition node; wherein the growth performance data includes at least one of body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth.

[0007] Optionally, the pedigree graph includes three layers of nodes; the first layer of nodes in the pedigree graph represents purebred pigs; the second layer of nodes in the pedigree graph represents two-way crossbred pigs and / or purebred pigs; wherein, the edge between the first layer node and the second layer node represents the pedigree relationship; the third layer of nodes in the pedigree graph represents the three-way crossbred pigs; the edge between the second layer node and the third layer node represents the pedigree relationship.

[0008] Optionally, the growth performance data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the method further includes: predicting the amino acid content of each three-way crossbred pig at the next acquisition node based on the growth performance data of each three-way crossbred pig at the next acquisition node; and optimizing the feed formulation based on the amino acid content of each three-way crossbred pig at the next acquisition node.

[0009] Optionally, predicting the amino acid content of each three-way crossbred pig at the next acquisition node based on the growth performance data of each three-way crossbred pig at the next acquisition node includes: performing multinomial regression based on sample data to construct a multinomial regression model; wherein, the sample data includes growth characteristic data of M sample three-way crossbred pigs and their corresponding amino acid contents; the multinomial regression model characterizes the correspondence between the growth performance data of the sample three-way crossbred pigs and the amino acid content of the sample three-way crossbred pigs; M is a positive integer; the growth performance data of each three-way crossbred pig at the next acquisition node is input into the multinomial regression model, and the predicted amino acid content of each three-way crossbred pig at the next acquisition node is output.

[0010] Optionally, before obtaining the growth characteristic data of N three-way crossbred pigs in the target pigpen, the method further includes: selecting 300 three-way crossbred pigs; wherein, the 300 three-way crossbred pigs are divided into half males and half females; randomly distributing the 300 three-way crossbred pigs into 10 pigpens; wherein, each pigpens are designated as the target pigpens; each pigpens are not fed, and the stocking density of each pigpens is A; A ranges from 1.1 to 1.3 cubic meters per pen; N is 30.

[0011] Optionally, acquiring the growth characteristic data of N three-way crossbred pigs in the target pigpen includes: collecting the weight of each three-way crossbred pig at preset intervals during the feeding period; collecting the total feed intake of the target pigpen at preset intervals during the feeding period, and calculating the feed intake of each three-way crossbred pig in the target pigpen based on the total feed intake; and collecting the backfat thickness at the 10th rib, the eye muscle area at the 10th rib, and the eye muscle depth at the 10th rib of each three-way crossbred pig at preset intervals during the feeding period.

[0012] Optionally, acquiring the growth characteristic data of N three-way crossbred pigs in the target pig pen includes: acquiring the growth characteristic data of the first Q collection nodes of the N three-way crossbred pigs in the target pig pen at the current time; where Q is a positive integer greater than or equal to 10; correspondingly, inputting the growth characteristic data of the N three-way crossbred pigs in the target pig pen and the pedigree map into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next collection node includes: inputting the growth characteristic data of the first Q collection nodes of the N three-way crossbred pigs in the target pig pen at the current time and the pedigree map into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next collection node.

[0013] Optionally, the method further includes: selecting breeding pigs for subsequent breeding based on the growth performance data of each three-way crossbred pig at the next acquisition node.

[0014] Optionally, the three-way crossbred pig is a Duroc pig, a Landrace pig, and a Yorkshire pig obtained through crossbreeding.

[0015] Secondly, this application also provides a growth performance prediction system for three-way crossbred pigs, comprising: an acquisition module for acquiring growth characteristic data of N three-way crossbred pigs in a target pigpen, and the pedigree of each three-way crossbred pig; wherein the growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the growth characteristic data are all periodically collected time-series data; N is a positive integer; a construction module for constructing a pedigree map based on the pedigree of each three-way crossbred pig; and a prediction module for inputting the growth characteristic data of the N three-way crossbred pigs in the target pigpen, and the pedigree map, into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next acquisition node; wherein the growth performance data includes at least one of body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth.

[0016] The beneficial effects of this invention include: First, traditional pig growth performance prediction neglects the genetic background of pigs. The technical solution of this application, however, in the growth prediction process for three-way crossbred pigs, generates a pedigree map by obtaining the bloodline relationship of each three-way crossbred pig, and then combines this pedigree map with the prediction of the growth performance of the three-way crossbred pigs. Since the genes of pigs directly affect the growth potential and health status of offspring during the crossbreeding process, the above process allows for a better understanding of the potential impact of genotype on the growth of three-way crossbred pigs, thereby improving the accuracy of growth performance prediction for three-way crossbred pigs.

[0017] Secondly, this application employs a growth performance prediction method based on multi-dimensional data. A single indicator is insufficient to reflect the complete production status of pigs. For example, existing technologies use weight data to predict growth performance, but this approach lacks rationality. In this application, however, the weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth of each three-way crossbred pig are periodically collected. This data, combined with the bloodline of each three-way crossbred pig, provides a multi-faceted and comprehensive predictive analysis (comprehensive assessment). In other words, the combined effect of multi-dimensional data further improves the accuracy of growth performance prediction for three-way crossbred pigs, enabling more precise predictions of their future growth performance. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the steps of a method for predicting the growth performance of three-way crossbred pigs provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a bloodline chart provided in an embodiment of the present invention; Figure 3 A flowchart illustrating the steps of another method for predicting the growth performance of three-way crossbred pigs provided in an embodiment of the present invention; Figure 4 A block diagram of a growth performance prediction system for three-way crossbred pigs provided for the implementation of this invention; Figure 5 This is a module block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0020] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] Since three-way crossbred pigs are obtained by crossing three different breeds of pigs, predicting their growth performance is extremely important. Predicting the growth performance of three-way crossbred pigs helps farms understand the growth potential of piglets in advance, select breeding pigs with excellent growth characteristics, and identify potential health problems, among other things.

[0022] However, research has found that existing predictions of pig growth performance are based solely on traditional empirical data and cannot be effectively applied to three-way crossbred pigs, resulting in very low accuracy in predicting the growth performance of three-way crossbred pigs.

[0023] In view of the above problems, this application proposes the following embodiments to solve the above technical problems.

[0024] Please see Figure 1 This application provides a method for predicting the growth performance of three-way crossbred pigs, including steps 101 to 103.

[0025] Step 101: Obtain the growth characteristic data of N three-way crossbred pigs in the target pigpen, as well as the blood relationship of each three-way crossbred pig.

[0026] The growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; all growth characteristic data are periodically collected time-series data. Understandably, during the rearing of three-way crossbred pigs, breeders can collect the above-mentioned growth characteristic data at preset intervals to obtain periodically collected time-series growth characteristic data. This time interval can be set according to actual needs, such as once a day, twice a day, five days, etc.

[0027] It should be noted that N is a positive integer, and the number of N represents the number of three-way crossbred pigs in the target pigpen.

[0028] Step 102: Construct a kinship map based on the bloodline of each three-way crossbred pig.

[0029] Then, based on the bloodline of each three-way crossbred pig, a phylogenetic map of the entire pig pen can be constructed for that target pig pen.

[0030] It should be noted that during the breeding process, the paternal and maternal parents of each three-way crossbred pig are recorded. Therefore, the bloodline of each three-way crossbred pig can be directly obtained to construct a bloodline map.

[0031] It is understandable that there is at least one overlapping relationship in this bloodline map, such as the father of three-way crossbred pig A being the same as the father of three-way crossbred pig B, and the mother of three-way crossbred pig A being the same as the mother of three-way crossbred pig C.

[0032] Step 103: Input the growth characteristic data and bloodline map of N three-way crossbred pigs in the target pigpen into the set dynamic graph convolution model, and output the growth performance data of each three-way crossbred pig in the next acquisition node.

[0033] Then, the growth characteristic data of N three-way crossbred pigs in the target pigpen are combined with the bloodline map, and the growth performance data of each three-way crossbred pig corresponding to the next acquisition node is output through the set dynamic graph convolution model.

[0034] The growth performance data include at least one of the following: body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth.

[0035] That is, in one embodiment, at least one of the following parameters for the next node—body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth—can be output using the method described above. In other embodiments, the above five production performance data can be output using the method described above.

[0036] In other words, the output growth performance data can be exactly the same as the input growth characteristic data, or the output growth performance data can be different from the input growth characteristic data.

[0037] In summary, the embodiments of this application provide a method for predicting the growth performance of three-way crossbred pigs, which has the following beneficial effects: First, traditional pig growth performance prediction neglects the genetic background of pigs. The technical solution of this application, however, in the growth prediction process for three-way crossbred pigs, generates a pedigree map by obtaining the bloodline relationship of each three-way crossbred pig, and then combines this pedigree map with the prediction of the growth performance of the three-way crossbred pigs. Since the genes of pigs directly affect the growth potential and health status of offspring during the crossbreeding process, the above process allows for a better understanding of the potential impact of genotype on the growth of three-way crossbred pigs, thereby improving the accuracy of growth performance prediction for three-way crossbred pigs.

[0038] Secondly, this application employs a growth performance prediction method based on multi-dimensional data. A single indicator is insufficient to reflect the complete production status of pigs. For example, existing technologies use weight data to predict growth performance, but this approach lacks rationality. In this application, however, the weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth of each three-way crossbred pig are periodically collected. This data, combined with the bloodline of each three-way crossbred pig, provides a multi-faceted and comprehensive predictive analysis (comprehensive assessment). In other words, the combined effect of multi-dimensional data further improves the accuracy of growth performance prediction for three-way crossbred pigs, enabling more precise predictions of their future growth performance.

[0039] Third, since growth characteristic data (such as body weight and feed intake) have strong temporal characteristics, meaning that these growth characteristic data change over time, this application provides a dynamic graph convolutional model to effectively capture these temporal data, enabling the model to capture the changing trends in the time dimension and thus provide effective and reasonable predictions.

[0040] Optionally, the pedigree graph includes three layers of nodes; the first layer of nodes in the pedigree graph represents purebred pigs; the second layer of nodes in the pedigree graph represents two-way crossbred pigs and / or purebred pigs; wherein, the edge between the first layer node and the second layer node represents the pedigree relationship; the third layer of nodes in the pedigree graph represents three-way crossbred pigs; the edge between the second layer node and the third layer node represents the pedigree relationship.

[0041] For example, please refer to Figure 2 , Figure 2 In the diagram, gray nodes represent boars, and white nodes represent sows. The first-level nodes (A, B, C) represent different purebred pigs, i.e., one-way crossbred pigs. The sire and dam in these first-level nodes can represent different breeds. The second-level nodes (D, E, F) represent different two-way crossbred pigs and / or purebred pigs. The sire and dam in these second-level nodes can also represent different breeds. The third-level nodes (G, H, I, J) represent different three-way crossbred pigs.

[0042] As can be seen, by providing the above-described pedigree map, the embodiments of this application can clearly express the pedigree relationships between different pigs, so as to better analyze the impact of different pedigree relationships on the growth performance of pigs, better capture gene effects, and thus improve the accuracy and interpretability of subsequent predictions.

[0043] Optionally, the three-way crossbred pig is obtained by crossbreeding Duroc, Landrace, and Yorkshire pigs. That is, the three-way crossbred pig provided in the embodiments of this application can be a DLY pig.

[0044] The following is a complete description of the growth performance prediction method for three-way crossbred pigs provided in the embodiments of this application.

[0045] First, the data collection process will be explained.

[0046] Optionally, before obtaining the growth characteristic data of N three-way crossbred pigs in the target pigpen, the method further includes: selecting 300 three-way crossbred pigs; wherein, the 300 three-way crossbred pigs are divided into half males and half females; randomly distributing the 300 three-way crossbred pigs into 10 pigpens; wherein, each pigpens are designated as target pigpens; each pigpens are not fed, and the stocking density of each pigpens is A; A ranges from 1.1 to 1.3 cubic meters per pen; N is 30.

[0047] Specifically, 300 healthy New Zealand DLY crossbred piglets of the same batch, aged 25 days, weaned, and born within 3 days of each other, can be selected from a two-way crossbred pig farm. The piglets should be half male and half female, with the males already castrated. They should be randomly assigned to 10 pig pens, with 30 piglets per pen (half male and half female), and the bloodline of each pig in each pen should be recorded. Each pig pen will be used as the target pig pen for prediction in this application.

[0048] The enclosures are equipped with solid floors, cement walls, free-flowing feed troughs, free-flowing water, and no feed control, ensuring a stocking density of 1.1 to 1.3 cubic meters per animal.

[0049] It should be noted that by selecting 300 three-way crossbred pigs and ensuring that boars and sows each comprised half, a broader representativeness of the sample could be guaranteed. Sex balance helps eliminate potential biases in growth performance caused by sex, ensuring that the predictive model can more accurately adapt to the performance of pigs of different sexes. These 300 pigs were randomly assigned to 10 different pens, each with potentially different environmental and management conditions, thus generating more diverse growth characteristic data.

[0050] Secondly, raising pigs under uncontrolled feed conditions means that the rearing environment and feed formulation in each pen are not specially intervened in. This helps in studying the growth performance of pigs in normal or natural environments, avoiding the influence of artificial intervention on the data. The stocking density in each pen is set at 1.1~1.3 cubic meters / pen, which ensures that the pigs grow in a relatively standard spatial environment. Stocking density has a significant impact on the growth rate and health status of pigs, and this density design ensures that the growth of pigs is not affected by overcrowding, while maintaining a certain level of environmental stress.

[0051] The bloodline relationships recorded for each pigpen can be expressed by the formula: Here, Indicates the number of the pigpen; Indicates the first Blood relations within a pigsty. Indicates the first The bloodline of the first three-way crossbred pig in each pen is determined by the bloodline of that pig, and so on. Indicates the first The bloodline of the 30th three-way crossbred pig in the pigpen.

[0052] Please see Figure 3 Optionally, the above steps obtain growth characteristic data of N three-way crossbred pigs in the target pigpen, including: steps 301 to 303.

[0053] Step 301: Collect the weight of each three-way crossbred pig during the feeding period at preset intervals.

[0054] Specifically, during the rearing period, the weight of the three-way crossbred pigs in each pen can be recorded, and can be expressed by the formula: Here, Indicates the number of the pigpen; Indicates the first The weight data corresponding to each pigpen is a vector; Indicates the first The first three-way crossbred pig in the pigpen was collected at the sampling point. The weight below; and so on, Indicates the first The 30th three-way crossbred pig in the pigpen was collected at the sampling point. The weight below; among them, ; This indicates the total feeding time (total feeding duration). This indicates the aforementioned preset interval.

[0055] Step 302: Collect the total feed intake of the target pig pen at preset intervals during the feeding period, and calculate the feed intake of each three-way crossbred pig in the target pig pen based on the total feed intake.

[0056] The total feed intake of each pigpen can be expressed by the formula: ; Indicates the first A pigsty at the data collection node The total feed intake of the pigs is given below, and the feed intake of each three-way crossbred pig in the pen can be expressed as: That is, it is obtained by taking the average.

[0057] Step 303: According to the preset interval, the backfat thickness at the 10th rib, the eye muscle area at the 10th rib, and the eye muscle depth at the 10th rib of each three-way crossbred pig are collected at intervals during the feeding period.

[0058] Specifically, during the rearing period, the backfat thickness at the 10th rib of each three-way crossbred pig in each pen can be measured and expressed by the formula: Here, Indicates the number of the pigpen; Indicates the first The backfat thickness at the 10th rib corresponding to each pigpen is a vector; Indicates the first The first three-way crossbred pig in the pigpen was collected at the sampling point. The back fat is thicker at the 10th rib below the waist; and so on. Indicates the first The 30th three-way crossbred pig in the pigpen was collected at the sampling point. Thick back fat at the 10th rib below; The meaning of "is explained in the preceding text, and the same applies to its appearance in the following text" can be found in the previous explanation.

[0059] During the rearing period, the area of ​​the eye muscle at the 10th rib of each three-way crossbred pig in each pen can be counted and expressed by the formula: Here, Indicates the number of the pigpen; Indicates the first The area of ​​the eye muscle at the 10th rib corresponding to each pigpen is a vector. Indicates the first The first three-way crossbred pig in the pigpen was collected at the sampling point. The area of ​​the extraocular muscles at the 10th rib below; and so on. Indicates the first The 30th three-way crossbred pig in the pigpen was collected at the sampling point. The area of ​​the extraocular muscles at the 10th rib below the rib.

[0060] During the rearing period, the depth of the oculomotor muscle at the 10th rib of each three-way crossbred pig in each pen can be measured and expressed by the formula: Here, Indicates the number of the pigpen; Indicates the first The depth of the extraocular muscle at the 10th rib corresponding to each pigpen is a vector. Indicates the first The first three-way crossbred pig in the pigpen was collected at the sampling point. The depth of the extraocular muscles is at the 10th rib below; and so on. Indicates the first The 30th three-way crossbred pig in the pigpen was collected at the sampling point. The depth of the extraocular muscles at the 10th rib below.

[0061] Among them, the backfat thickness and eye muscle area can be obtained by referring to the B-mode ultrasound method for measuring the backfat thickness and eye muscle area of ​​live pigs.

[0062] It should be noted that the backfat thickness, eye muscle area, and eye muscle depth at the 10th rib are used in this application embodiment primarily because the 10th rib is a relatively stable location within the pig and a common, standardized measurement site. It is not only easy to locate in different pigs but also consistent across different breeds, sexes, and weight stages. Muscle and fat accumulation at the 10th rib typically reflects the pig's growth and development status. The backfat thickness, eye muscle area, and eye muscle depth at this location are strongly correlated with the overall fat and muscle distribution of the pig, thus effectively reflecting the pig's overall growth. Furthermore, the 10th rib is relatively easy to measure.

[0063] Additionally, it should be noted that if any pigs die during the rearing period, the area of ​​the rearing area should be reduced accordingly to ensure that the stocking density remains at 1.1 to 1.3 cubic meters per pig.

[0064] The dynamic graph convolution model provided in the embodiments of this application is described below. First, the growth characteristic data of three-way crossbred pigs are organized into a formula for each pigpen: ; represents the growth characteristic data of all pigs in the first pigpen, such as Specifically :in, The relevant parameters in this formula can be found in the explanations in the foregoing embodiments.

[0065] Then, The elements in the graph serve as the values ​​of the third-level nodes in the kinship diagram of each pigpen. For example, the value of the first node in the third level of the first pen is... The values ​​of the nodes in the second and first layers of the kinship graph are set to a length equal to... A vector, such as a second-level node, can be represented as... The third-level node can be represented as All of them are learnable vectors.

[0066] Then, dynamic graph convolution can be performed on each pigpen.

[0067] The expression for the dynamic graph convolution model can be: ; in, Represents a node The first in The output value after performing operations on a tensor Indicates the pig's serial number. Represents the convolution weights. Indicates intermediate variables. Represents a node The first in Tensors; and Respectively represent and Connected and The first in Values.

[0068] After completing the calculation using the above formula, you can take... time The linear layer (Line) of the model outputs the growth performance data of the three-way crossbred pig at the next data collection node, as shown below: .

[0069] Optionally, the above steps for obtaining growth characteristic data of N three-way crossbred pigs in the target pig pen include: obtaining growth characteristic data of the first Q collection nodes of the N three-way crossbred pigs in the target pig pen at the current time; where Q is a positive integer greater than or equal to 10; correspondingly, the above steps input the growth characteristic data of the N three-way crossbred pigs in the target pig pen and the pedigree map into a set dynamic graph convolution model, and output the growth performance data of each three-way crossbred pig at the next collection node, including: inputting the growth characteristic data of the first Q collection nodes of the N three-way crossbred pigs in the target pig pen at the current time and the pedigree map into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next collection node.

[0070] That is, in the above process, data from any Q historical collection nodes can be used to predict the growth performance data of each three-way crossbred pig at the next collection node.

[0071] Optionally, growth performance data include body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the method also includes: predicting the amino acid content of each three-way crossbred pig at the next acquisition node based on the growth performance data of each three-way crossbred pig at the next acquisition node; and optimizing the feed formulation based on the amino acid content of each three-way crossbred pig at the next acquisition node.

[0072] It should be noted that amino acids are essential nutrients for the growth and development of three-way crossbred pigs. Their requirements are influenced by various factors such as breed, sex, growth stage, and environmental factors. Therefore, accurately predicting the amino acid nutritional requirements of three-way crossbred pigs is crucial for optimizing diet formulation and improving growth performance. In this application, growth performance data of each three-way crossbred pig at the next data collection point can be used to predict the amino acid content of each three-way crossbred pig at the next data collection point, thereby optimizing feed formulation.

[0073] Optionally, based on the growth performance data of each three-way crossbred pig at the next acquisition node, the amino acid content of each three-way crossbred pig at the next acquisition node is predicted, including: performing multinomial regression based on sample data to construct a multinomial regression model; wherein, the sample data includes the growth characteristic data of M sample three-way crossbred pigs and the corresponding amino acid content; the multinomial regression model characterizes the correspondence between the growth performance data of the sample three-way crossbred pigs and the amino acid content of the sample three-way crossbred pigs; M is a positive integer; the growth performance data of each three-way crossbred pig at the next acquisition node is input into the multinomial regression model, and the predicted amino acid content of each three-way crossbred pig at the next acquisition node is output.

[0074] The expression for a multinomial regression model can be: ; In this method, Indicates the first The first pigsty The amino acid content of three-way crossbred pigs. , , , , , , , , as well as All of these are polynomial parameters obtained by performing polynomial regression on the sample data.

[0075] The aforementioned optimized feed formulation can be characterized as follows: if the amino acid content is higher than that that can be obtained through normal feeding, additional supplements can be used to ensure the normal growth of pigs.

[0076] Optionally, the method further includes: selecting breeding pigs for subsequent breeding based on the growth performance data of each three-way crossbred pig at the next acquisition node.

[0077] Specifically, based on the prediction results, the growth of three-way crossbred pigs in each pigpen is statistically predicted, and the bloodline of the pigpen with the best growth is selected as the recommended plan for subsequent piglet breeding (the breeds of pigs in the second and first layers of the bloodline map are known, so a bloodline of the best-growing piglets can also be selected for subsequent breeding).

[0078] As can be seen, the prediction method provided in this application embodiment can further screen breeding pigs for subsequent reproduction. This method can provide excellent breeding pig breeding efficiency, accelerate genetic improvement, and improve the production efficiency of pig herds.

[0079] Furthermore, in one embodiment, the dynamic graph convolution model provided in this application can be configured with different models according to different growth stages of three-way crossbred pigs.

[0080] It should be noted that the growth of three-way crossbred pigs exhibits different characteristics and growth rates at different stages, and the growth factors differ at each stage. Therefore, in this embodiment, sample three-way crossbred pigs at different stages can be used to train different dynamic graph convolutional models. This further allows the method to include: determining N growth stages of the three-way crossbred pigs in the current target pigpen; then, inputting the growth characteristic data and pedigree maps of the N three-way crossbred pigs in the target pigpen into a dynamic graph convolutional model that matches the growth stages of the N three-way crossbred pigs in the current target pigpen; and outputting the growth performance data of each three-way crossbred pig at the next acquisition node.

[0081] For example, it can be divided into three growth stages: Initial stage: The weight of the three-way crossbred pigs is below 20 kg.

[0082] Mid-stage: The weight of the three-way crossbred pigs is between 20 kg and 50 kg.

[0083] Later stage: The weight of the three-way crossbred pigs is over 50 kg.

[0084] The three growth stages mentioned above correspond to three different pre-trained dynamic graph convolutional models. By predicting growth at these multiple stages and combining time series data, pedigree maps, and staged modeling, the changing patterns of pigs at different growth stages can be accurately captured. Furthermore, the results are optimized through a multi-stage joint prediction mechanism, improving the accuracy and stability of the predictions.

[0085] Please see Figure 4 Based on the same inventive concept, this application also provides a growth performance prediction system 400 for three-way crossbred pigs, comprising: an acquisition module 401, used to acquire growth characteristic data of N three-way crossbred pigs in a target pigpen, and the blood relationship of each three-way crossbred pig; wherein, the growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the growth characteristic data are all periodically collected time-series data; N is a positive integer; a construction module 402, used to construct a blood relationship map based on the blood relationship of each three-way crossbred pig; and a prediction module 403, used to input the growth characteristic data of the N three-way crossbred pigs in the target pigpen, and the blood relationship map, into a set dynamic graph convolution model, and output the growth performance data of each three-way crossbred pig in the next acquisition node; wherein, the growth performance data includes at least one of body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth.

[0086] Please see Figure 5 Based on the same inventive concept, this application provides a module frame for an electronic device 500 that applies the above-described method. The electronic device 500 includes: at least one processor 501 (… Figure 5 (Only one is shown in the image), memory 502, computer program 503 stored in memory 502 and executable on at least one processor 501, processor 501 executing computer program 503 to implement the steps of the method in any of the foregoing embodiments.

[0087] The electronic device 500 can be a server, a personal computer, a laptop, etc.

[0088] Those skilled in the art will understand that Figure 5 This is merely an example of electronic device 500 and does not constitute a limitation on electronic device 500. It may include more or fewer components than shown, or combine certain components, or use different components.

[0089] The processor 501 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0090] In some embodiments, the memory 502 may be an internal storage unit of the electronic device 500, such as a hard disk or memory of the electronic device 500. In other embodiments, the memory 502 may be an external storage device of the electronic device 500, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 500. Furthermore, the memory 502 may include both internal storage units and external storage devices of the electronic device 500.

[0091] It should be noted that the above-mentioned systems, devices, etc. are based on the same concept as the method embodiments of this application. The modules designed in the system, as well as the steps performed by the device and the resulting technical effects, can all be found in the method embodiments section, and will not be repeated here.

[0092] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0093] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0094] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0095] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0096] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0097] Those skilled in the art will recognize that the units and algorithm steps of the various examples 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method of predicting growth performance of a three-way cross pig, characterized in that, include: Acquire growth characteristic data of N three-way crossbred pigs in the target pigpen, as well as the blood relationship of each three-way crossbred pig; wherein, the growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the growth characteristic data are all time-series data collected periodically; N is a positive integer; A kinship map is constructed based on the bloodline relationship of each of the three-way crossbred pigs; The growth characteristic data of N three-way crossbred pigs in the target pigpen, as well as the bloodline map, are input into a set dynamic graph convolution model, and the growth performance data of each three-way crossbred pig in the next acquisition node are output; wherein, the growth performance data includes at least one of body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth.

2. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, The kinship map includes three layers of nodes; The first layer of nodes in the bloodline diagram represents purebred pigs; The second-layer nodes in the kinship graph represent binary crossbred pigs and / or purebred pigs; wherein, the edges between the first-layer nodes and the second-layer nodes represent kinship relationships. The third-layer node in the kinship graph represents the three-way crossbred pig; the edge between the second-layer node and the third-layer node represents the kinship relationship.

3. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, The growth performance data include body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the method further includes: Based on the growth performance data of each three-way crossbred pig at the next acquisition node, predict the amino acid content of each three-way crossbred pig at the next acquisition node; The feed formulation is optimized based on the amino acid content of each three-way crossbred pig at the next data collection node.

4. The method of growth performance prediction of a three-way cross pig according to claim 3, characterized in that, The prediction of the amino acid content of each three-way crossbred pig at the next acquisition node, based on the growth performance data of each three-way crossbred pig at the next acquisition node, includes: A multinomial regression model is constructed based on the sample data; wherein, the sample data includes the growth characteristic data and corresponding amino acid content of M sample three-way crossbred pigs; the multinomial regression model characterizes the correspondence between the growth performance data of the sample three-way crossbred pigs and the amino acid content of the sample three-way crossbred pigs; M is a positive integer; The growth performance data of each three-way crossbred pig at the next acquisition node is input into the polynomial regression model, and the predicted amino acid content of each three-way crossbred pig at the next acquisition node is output.

5. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, Before acquiring the growth characteristic data of N three-way crossbred pigs in the target pigpen, the method further includes: 300 three-way crossbred pigs were selected; among them, boars and sows each accounted for half of the 300 three-way crossbred pigs; The 300 three-way crossbred pigs were randomly assigned to 10 pig pens; each pig pen was designated as the target pig pen; feed was not controlled in each pig pen, and the stocking density in each pig pen was A. The range of A is 1.1~1.3 cubic meters / unit; N is 30.

6. The method of growth performance prediction of a three-way cross pig according to claim 5, characterized in that, The acquisition of growth characteristic data of N three-way crossbred pigs in the target pigpen includes: The body weight of each three-way crossbred pig was collected at preset intervals during the feeding period; According to the preset interval, the total feed intake of the target pig pen during the feeding period is collected at intervals, and the feed intake of each three-way crossbred pig in the target pig pen is calculated based on the total feed intake. According to the preset interval, the backfat thickness at the 10th rib, the eye muscle area at the 10th rib, and the eye muscle depth at the 10th rib of each three-way crossbred pig were collected at intervals during the feeding period.

7. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, The acquisition of growth characteristic data of N three-way crossbred pigs in the target pigpen includes: Obtain the growth characteristic data of the first Q collection nodes of N three-way crossbred pigs in the target pig herd at the current time; where Q is a positive integer greater than or equal to 10; Accordingly, the step of inputting the growth characteristic data of N three-way crossbred pigs in the target pigpen and the pedigree map into a set dynamic graph convolution model, and outputting the growth performance data of each three-way crossbred pig at the next acquisition node, includes: The growth characteristic data of the first Q collection nodes of the N three-way crossbred pigs of the target pig herd at the current time, as well as the blood relationship map, are input into the set dynamic graph convolution model, and the growth performance data of each three-way crossbred pig at the next collection node are output.

8. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, The method further includes: Based on the growth performance data of each three-way crossbred pig at the next data collection node, breeding pigs are selected for subsequent reproduction.

9. The method of growth performance prediction of a three-way cross pig according to claim 1, characterized in that, The three-way crossbred pigs are bred from Duroc, Landrace, and Yorkshire pigs through crossbreeding.

10. A growth performance prediction system for three-way crossbred pigs, characterized in that, include: The acquisition module is used to acquire growth characteristic data of N three-way crossbred pigs in the target pigpen, as well as the blood relationship of each three-way crossbred pig; wherein, the growth characteristic data includes body weight, feed intake, backfat thickness, eye muscle area, and eye muscle depth; the growth characteristic data are all time-series data collected periodically; N is a positive integer; A construction module is used to construct a kinship map based on the blood relationship of each of the three-way crossbred pigs; The prediction module is used to input the growth characteristic data of N three-way crossbred pigs in the target pigpen and the blood relationship map into a set dynamic graph convolution model, and output the growth performance data of each three-way crossbred pig in the next acquisition node; wherein, the growth performance data includes at least one of body weight, feed intake, backfat thickness, eye muscle area and eye muscle depth.

Images