Method and system for predicting poultry growth model based on environmental monitoring

Abstract

The invention provides a method and a system for predicting a poultry growth model based on environmental monitoring. The method comprises the following steps: acquiring a historical image and a current image of each breeding cage in which poultry is bred; training a support vector machine model according to the historical images to obtain a trained support vector machine model, obtaining first future growth index data corresponding to each current image according to the trained support vector machine model, and calculating second future growth index data corresponding to each current image; and performing weighted summation on the first future growth index data and the second future growth index data to obtain final future growth index data corresponding to each current image. According to the method, environmental monitoring data is introduced into the poultry breeding field, a prediction model is constructed by using a machine learning algorithm, the limitation of a traditional breeding method is broken through, a novel and scientific growth index prediction method is provided, and subjectivity and inaccuracy of traditional manual judgment are reduced.

Description

Technical Field

[0001] The present invention relates to the field of agricultural technology, and in particular to a prediction method and system for a poultry growth model based on environmental monitoring. Background Art

[0002] In traditional poultry farming, the prediction of poultry growth indicators mainly relies on experience and manual judgment, which is subjective and inaccurate. Therefore, a prediction model based on environmental monitoring is needed, which can scientifically predict poultry growth indicators through monitoring and analysis of environmental factors, and then provide scientific decision-making basis for breeders. Summary of the invention

[0003] The object of the present invention is to provide a prediction method and system for a poultry growth model based on environmental monitoring to improve the above-mentioned problems.

[0004] In order to achieve the above objectives, the present application provides the following technical solutions:

[0005] On the one hand, an embodiment of the present application provides a prediction method for a poultry growth model based on environmental monitoring, the method comprising:

[0006] Obtain historical and current images of each breeding cage in which poultry are raised;

[0007] Performing cluster analysis on all historical images to obtain at least two first clusters, screening all historical images according to the first clusters, clustering each remaining historical image again to obtain multiple second clusters, completing information annotation of each remaining historical image based on the second clusters, wherein the annotated information includes future growth indicator data; training a support vector machine model using the annotated historical images as training data to obtain a trained support vector machine model;

[0008] Inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image;

[0009] Perform cluster analysis on all current images to obtain a third cluster cluster, perform comparative analysis on each third cluster cluster and each second cluster cluster, add each current image to the second cluster cluster according to the comparative analysis result, and obtain the second future growth indicator data corresponding to each current image; perform weighted summation on the first future growth indicator data and the second future growth indicator data, and obtain the final future growth indicator data corresponding to each current image.

[0010] In a second aspect, an embodiment of the present application provides a prediction system for a poultry growth model based on environmental monitoring, the system comprising:

[0011] An acquisition module, used for acquiring historical images and current images of each breeding cage where poultry are raised;

[0012] A training module, used to perform cluster analysis on all historical images to obtain at least two first clusters, screen all historical images according to the first clusters, cluster each remaining historical image again to obtain multiple second clusters, complete information annotation of each remaining historical image based on the second clusters, and the annotate the information including future growth indicator data; train the support vector machine model using the annotated historical images as training data to obtain a trained support vector machine model;

[0013] A first calculation module, used for inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image;

[0014] The second calculation module is used to perform cluster analysis on all current images to obtain third cluster clusters, compare and analyze each third cluster cluster with each second cluster cluster, add each current image to the second cluster cluster according to the comparison and analysis results, and obtain second future growth indicator data corresponding to each current image; perform weighted summation on the first future growth indicator data and the second future growth indicator data to obtain final future growth indicator data corresponding to each current image.

[0015] In a third aspect, an embodiment of the present application provides a prediction device for a poultry growth model based on environmental monitoring, the device comprising a memory and a processor. The memory is used to store a computer program; the processor is used to implement the steps of the above-mentioned prediction method for a poultry growth model based on environmental monitoring when executing the computer program.

[0016] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for predicting the poultry growth model based on environmental monitoring are implemented.

[0017] The beneficial effects of the present invention are:

[0018] In the present invention, the support vector machine model is trained using historical images, and then the trained support vector machine model is used to predict the first future growth index data of the current image, wherein, during the training, the original historical images are screened, and then a fast annotation method is used for them, by which the annotation speed can be improved, and then the training speed can be improved; at the same time, considering that the prediction of multiple dimensions may make the final prediction result more accurate, the present invention also clusters and similarity analyzes the current image, and attributes it to different second clustering clusters, by which the second future growth index data corresponding to the current image can be obtained, and finally the weighted sum can be calculated to obtain the final future growth index data corresponding to each current image. The present invention introduces environmental monitoring data in the field of poultry farming, and uses machine learning algorithms to construct a prediction model, breaking through the limitations of traditional farming methods, providing a new and scientific growth index prediction method, and reducing the subjectivity and inaccuracy of traditional manual judgment. And by predicting the future growth index data of poultry, breeders can make timely adjustments according to actual conditions and environmental changes, optimize feeding management and input, and improve farming efficiency.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 It is a schematic flow chart of a prediction method of a poultry growth model based on environmental monitoring according to an embodiment of the present invention;

[0022] Figure 2 Schematic diagram of the prediction system structure of the poultry growth model based on environmental monitoring described in an embodiment of the present invention;

[0023] Figure 3 It is a schematic diagram of the structure of the prediction device of the poultry growth model based on environmental monitoring described in the embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals or letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0026] Example 1

[0027] like Figure 1 As shown, this embodiment provides a prediction method for a poultry growth model based on environmental monitoring, which includes step S1, step S2, step S3 and step S4.

[0028] Step S1, obtaining historical images and current images of each breeding cage where poultry are raised;

[0029] This step can be understood as, for example, if there are chickens in a chicken cage, then an image of the chicken cage is obtained, and the image includes the chicken cage and the chickens. The poultry in this embodiment is an oviparous animal, and the poultry in the same breeding cage can be hatched at the same time, for example, on the same day. The historical image can be understood as an image obtained in the past year, and the current image can be understood as an image of the day.

[0030] Step S2, clustering all historical images to obtain at least two first clusters, screening all historical images according to the first clusters, clustering each remaining historical image again to obtain multiple second clusters, completing information annotation of each remaining historical image based on the second clusters, the annotated information including future growth indicator data; training the support vector machine model using the annotated historical images as training data to obtain a trained support vector machine model;

[0031] In this step, all historical images are first screened, the purpose of which is to improve the accuracy of subsequent model training. The specific implementation steps of this step include step S21 and step S22;

[0032] Step S21, analyzing each of the first clusters, wherein each of the first clusters contains at least two of the historical images, counting the number of historical image pairs in the first cluster, recorded as a first value A, wherein the historical image pair is composed of any two different historical images; calculating a first similarity between two historical images in each historical image pair, comparing the first similarity with a preset first similarity threshold, counting the number of first similarities that do not exceed the first similarity threshold, recorded as a second value B; counting the number of historical images contained in the first cluster, recorded as a third value H;

[0033] In this step, the image pair can be understood as If the first similarity between the two images in the image pair does not exceed the first similarity threshold, then the second value B is increased by 1;

[0034] Step S22, complete the screening of all historical images according to the first numerical value A, the second numerical value B and the third numerical value H, record each remaining historical image as a remaining historical image, annotate the remaining historical images with information, and use all the remaining historical images after annotation to perform preliminary training on the support vector machine model to obtain a preliminary support vector machine model; input a part of the remaining historical images into the preliminary support vector machine model to obtain the predicted annotation information corresponding to the remaining historical images, send the predicted annotation information corresponding to the remaining historical images to the staff so that the staff can correct the predicted annotation information, and input the remaining historical images after the corrected predicted annotation information into the preliminary support vector machine model again for training to obtain the trained support vector machine model.

[0035] In this step, all historical images are screened according to the first value A, the second value B and the third value H, and each remaining historical image is recorded as a remaining historical image. The specific implementation steps of labeling the remaining historical images include step S221;

[0036] Step S221: Determine Is the value greater than If the value of is greater than, all historical images in this first cluster are deleted; after each first cluster is analyzed, each remaining historical image is recorded as a remaining historical image; all the remaining historical images are clustered again to obtain multiple second clusters; for each second cluster, a preset number of remaining historical images are screened out according to the distance from the center of the second cluster to form a labeled image set, and the labeled image set is sent to the staff for information labeling to obtain the labeling information corresponding to the labeled image set, and the labeling information corresponding to the labeled image set is used as the labeling information corresponding to each of the remaining historical images in the second cluster.

[0037] In this step, abnormal historical images can be deleted through screening operations; at the same time, the preset number can be customized. The purpose of forming a labeled image set is to consider that a single image may not be conducive to accurate judgment and annotation by the labeler, so multiple images are formed into a labeled image set and sent to the labeler; after sending, the labeler will conduct a comprehensive analysis of all images in the labeled image set, and after the comprehensive analysis, a future growth indicator data will be annotated for this labeled image set, for example, the survival rate in the future period can be predicted based on the density of the image; after obtaining the labeling information corresponding to this labeled image set, each remaining historical image in the second cluster corresponding to this labeled image set is annotated with this labeling information; This labeling method can quickly complete the labeling of all remaining historical images, thereby improving the labeling speed; however, although this labeling method is fast, the labeling accuracy is average. Therefore, when training in step S22, the support vector machine model is first preliminarily trained with the labeled historical images to obtain a preliminary support vector machine model; then a portion (customizable, such as one quarter) of the remaining historical images is input into the preliminary support vector machine model to obtain the predicted labeling information corresponding to the remaining historical images, which is then corrected and the preliminary support vector machine model is trained again. This simple and low-cost method can obtain a certain amount of more accurate training data, thereby further improving the training accuracy of the model.

[0038] Step S3, inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image;

[0039] Through the above steps, a trained support vector machine model can be obtained, and the first future growth indicator data corresponding to each current image can be obtained by inputting the current image into the trained support vector machine model;

[0040] Step S4: clustering all current images to obtain a third cluster, comparing and analyzing each third cluster with each second cluster, adding each current image to the second cluster according to the comparative analysis result to obtain second future growth indicator data corresponding to each current image; performing weighted summation on the first future growth indicator data and the second future growth indicator data to obtain final future growth indicator data corresponding to each current image.

[0041] In this step, the weight of the first future growth indicator data is 60%, and the weight of the second future growth indicator data is 40%; each third cluster is compared and analyzed with each second cluster, and each current image is added to the second cluster according to the comparison and analysis result, and the specific implementation steps of obtaining the second future growth indicator data corresponding to each current image include step S41 and step S42;

[0042] Step S41: for each third cluster, a feature vector of the cluster center image corresponding to the second cluster is calculated with a feature vector of the cluster center image corresponding to each second cluster to obtain a plurality of second similarities; when the second similarity is greater than a preset second similarity threshold, the corresponding second cluster is used as a similar cluster of each third cluster;

[0043] In this step, a trained CNN convolutional neural network model can be used to extract the feature vector of the image; or other feature extraction methods can be used, which are not limited in this step; after similarity calculation, there may be one or more corresponding similar clusters; therefore, step S42 needs to be executed; in this step, the second similarity threshold can be customized, but it is necessary to ensure that there is at least one similar cluster;

[0044] Step S42, determine the number of similar clustering clusters. If the number of similar clustering clusters is one, directly define this similar clustering cluster as the most similar clustering cluster corresponding to the third clustering cluster; otherwise, calculate the distance between each third clustering cluster and each corresponding similar clustering cluster, and screen out the most similar clustering cluster corresponding to each third clustering cluster according to the distance calculation result; add the current image in each third clustering cluster to its corresponding most similar clustering cluster, and use the annotation information corresponding to the most similar clustering cluster as the annotation information of the current image added to the most similar clustering cluster, so as to obtain the second future growth indicator data corresponding to each current image.

[0045] By using the method of this step, the current image can be quickly assigned to different second clusters, and then the second future growth indicator data corresponding to each current image can be quickly obtained; at the same time, in this step, the distance between each third cluster and each corresponding similar cluster is calculated, and the specific implementation steps of selecting the most similar cluster corresponding to each third cluster according to the distance calculation result include step S421;

[0046] Step S421, calculating the distance between each third cluster and each corresponding similar cluster, wherein a current image is arbitrarily selected from the third cluster, and the distance between the selected current image and each historical image belonging to the similar cluster is calculated, and all distances are averaged and the average is used as the distance between the third cluster and the corresponding similar cluster; the similar cluster corresponding to the smallest distance is used as the most similar cluster corresponding to the third cluster.

[0047] In this embodiment, the support vector machine model is trained using historical images, and then the trained support vector machine model is used to predict the first future growth index data of the current image. During the training, the original historical images are screened, and then a fast annotation method is used for them. This method can improve the annotation speed, thereby improving the training speed. At the same time, considering that the prediction of multiple dimensions may make the final prediction result more accurate, this embodiment also clusters and analyzes the current image, and attributes it to different second clustering clusters. Through this method, the second future growth index data corresponding to the current image can be obtained, and finally the weighted sum can be calculated to obtain the final future growth index data corresponding to each current image. This embodiment introduces environmental monitoring data in the field of poultry farming, and uses machine learning algorithms to build a prediction model, breaking through the limitations of traditional farming methods, providing a new and scientific growth index prediction method, and reducing the subjectivity and inaccuracy of traditional manual judgment. And by predicting the future growth index data of poultry, breeders can make timely adjustments according to actual conditions and environmental changes, optimize feeding management and input, and improve farming efficiency.

[0048] Example 2

[0049] like Figure 2 As shown, this embodiment provides a prediction system for poultry growth model based on environmental monitoring, and the system includes an acquisition module 1, a training module 2, a first calculation module 3 and a second calculation module 4.

[0050] An acquisition module 1 is used to acquire historical images and current images of each breeding cage where poultry are raised;

[0051] Training module 2 is used to perform cluster analysis on all historical images to obtain at least two first clusters, screen all historical images according to the first clusters, cluster each remaining historical image again to obtain multiple second clusters, complete information annotation of each remaining historical image based on the second clusters, and the annotate the information including future growth indicator data; use the annotated historical images as training data to train the support vector machine model to obtain a trained support vector machine model;

[0052] A first calculation module 3, used for inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image;

[0053] The second calculation module 4 is used to perform cluster analysis on all current images to obtain third cluster clusters, compare and analyze each third cluster cluster with each second cluster cluster, add each current image to the second cluster cluster according to the comparison and analysis results, and obtain the second future growth indicator data corresponding to each current image; perform weighted summation on the first future growth indicator data and the second future growth indicator data to obtain the final future growth indicator data corresponding to each current image.

[0054] In a specific implementation of the present disclosure, the training module 2 further includes a first calculation unit 21 and a training unit 22 .

[0055] The first calculation unit 21 is used to analyze each of the first clusters, wherein each of the first clusters contains at least two of the historical images, and count the number of historical image pairs in the first cluster, which is recorded as a first value A, and the historical image pair is composed of any two different historical images; calculate the first similarity of the two historical images in each historical image pair, compare the first similarity with a preset first similarity threshold, and count the number of first similarities that do not exceed the first similarity threshold, which is recorded as a second value B; count the number of historical images contained in the first cluster, which is recorded as a third value C;

[0056] The training unit 22 is used to complete the screening of all historical images according to the first numerical value A, the second numerical value B and the third numerical value C, record each remaining historical image as a remaining historical image, annotate the remaining historical images with information, and use all the remaining historical images after annotation to perform preliminary training on the support vector machine model to obtain a preliminary support vector machine model; input a part of the remaining historical images into the preliminary support vector machine model to obtain the predicted annotation information corresponding to the remaining historical images, send the predicted annotation information corresponding to the remaining historical images to the staff so that the staff can correct the predicted annotation information, and input the remaining historical images after the corrected predicted annotation information into the preliminary support vector machine model again for training to obtain a trained support vector machine model.

[0057] In a specific implementation of the present disclosure, the training unit 22 further includes a labeling unit 221 .

[0058] The marking unit 221 is used to determine Is the value greater than If the value of is greater than, all historical images in this first cluster are deleted; after each first cluster is analyzed, each remaining historical image is recorded as a remaining historical image; all the remaining historical images are clustered again to obtain multiple second clusters; for each second cluster, a preset number of remaining historical images are screened out according to the distance from the center of the second cluster to form a labeled image set, and the labeled image set is sent to the staff for information labeling to obtain the labeling information corresponding to the labeled image set, and the labeling information corresponding to the labeled image set is used as the labeling information corresponding to each of the remaining historical images in the second cluster.

[0059] In a specific implementation of the present disclosure, the second calculation module 4 further includes a second calculation unit 41 and a third calculation unit 42 .

[0060] A second calculation unit 41 is used to calculate the similarity between the feature vector of the cluster center image corresponding to the second cluster and the feature vector of the cluster center image corresponding to each second cluster for each third cluster, to obtain multiple second similarities, and when the second similarity is greater than a preset second similarity threshold, the corresponding second cluster is used as a similar cluster of each third cluster;

[0061] The third calculation unit 42 is used to determine the number of similar clusters. If the number of similar clusters is one, this similar cluster is directly defined as the most similar cluster corresponding to the third cluster; otherwise, the distance between each third cluster and each corresponding similar cluster is calculated, and the most similar cluster corresponding to each third cluster is screened out according to the distance calculation result; the current image in each third cluster is added to the corresponding most similar cluster, and the annotation information corresponding to the most similar cluster is used as the annotation information of the current image added to the most similar cluster, so as to obtain the second future growth indicator data corresponding to each current image.

[0062] In a specific implementation of the present disclosure, the third calculation unit 42 further includes a fourth calculation unit 421 .

[0063] The fourth calculation unit 421 is used to calculate the distance between each third clustering cluster and each corresponding similar clustering cluster, wherein a current image is arbitrarily selected from the third clustering cluster, and the distance between the selected current image and each historical image belonging to the similar clustering cluster is calculated, and the mean of all distances is calculated, and the mean is used as the distance between the third clustering cluster and the corresponding similar clustering cluster; the similar clustering cluster corresponding to the smallest distance is used as the most similar clustering cluster corresponding to the third clustering cluster.

[0064] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0065] Example 3

[0066] Corresponding to the above method embodiments, the embodiments of the present disclosure also provide a prediction device for a poultry growth model based on environmental monitoring. The prediction device for a poultry growth model based on environmental monitoring described below and the prediction method for a poultry growth model based on environmental monitoring described above can be referenced to each other.

[0067] Figure 3 FIG. 3 is a block diagram of a prediction device 300 for a poultry growth model based on environmental monitoring according to an exemplary embodiment. Figure 3 As shown, the prediction device 300 of the poultry growth model based on environmental monitoring may include: a processor 301 and a memory 302. The prediction device 300 of the poultry growth model based on environmental monitoring may also include one or more of a multimedia component 303, an I / O interface 304, and a communication component 305.

[0068] The processor 301 is used to control the overall operation of the prediction device 300 based on the environment monitoring poultry growth model to complete all or part of the steps in the prediction method based on the environment monitoring poultry growth model. The memory 302 is used to store various types of data to support the operation of the prediction device 300 based on the environment monitoring poultry growth model, such as instructions for any application or method operating on the prediction device 300 based on the environment monitoring poultry growth model, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 303 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 302 or sent via the communication component 305. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 304 provides an interface between the processor 301 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 305 is used for wired or wireless communication between the prediction device 300 of the poultry growth model based on environmental monitoring and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 305 can include: Wi-Fi module, Bluetooth module, NFC module.

[0069] In an exemplary embodiment, the prediction device 300 of the poultry growth model based on environmental monitoring can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned prediction method of the poultry growth model based on environmental monitoring.

[0070] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned prediction method of the poultry growth model based on environmental monitoring are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 302 including program instructions, and the above-mentioned program instructions can be executed by the processor 301 of the prediction device 300 for the poultry growth model based on environmental monitoring to complete the above-mentioned prediction method of the poultry growth model based on environmental monitoring.

[0071] Example 4

[0072] Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a readable storage medium. The readable storage medium described below and the prediction method of the poultry growth model based on environmental monitoring described above can refer to each other.

[0073] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the prediction method of the poultry growth model based on environmental monitoring in the above method embodiment.

[0074] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A prediction method for a poultry growth model based on environmental monitoring, characterized in that: include: Obtain historical and current images of each breeding cage in which poultry are raised; Performing cluster analysis on all historical images to obtain at least two first clusters, screening all historical images according to the first clusters, clustering each remaining historical image again to obtain multiple second clusters, completing information annotation of each remaining historical image based on the second clusters, wherein the annotated information includes future growth indicator data; training a support vector machine model using the annotated historical images as training data to obtain a trained support vector machine model; Inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image; Perform cluster analysis on all current images to obtain a third cluster cluster, perform comparative analysis on each third cluster cluster and each second cluster cluster, add each current image to the second cluster cluster according to the comparative analysis result, and obtain the second future growth indicator data corresponding to each current image; perform weighted summation on the first future growth indicator data and the second future growth indicator data, and obtain the final future growth indicator data corresponding to each current image.

2. The prediction method of the poultry growth model based on environmental monitoring according to claim 1, characterized in that: Performing cluster analysis on all historical images to obtain at least two first clusters, screening all historical images according to the first clusters, clustering each remaining historical image again to obtain multiple second clusters, and completing information annotation of each remaining historical image based on the second clusters, wherein the annotated information includes future growth indicator data; The annotated historical images are used as training data to train the support vector machine model, and the trained support vector machine model is obtained, including: Analyze each of the first clusters, wherein each of the first clusters contains at least two of the historical images, count the number of historical image pairs in the first cluster, and record it as a first value A, wherein the historical image pair is composed of any two different historical images; calculate the first similarity of the two historical images in each historical image pair, compare the first similarity with a preset first similarity threshold, and count the number of first similarities that do not exceed the first similarity threshold, and record it as a second value B; count the number of historical images contained in the first cluster, and record it as a third value H; Complete the screening of all historical images according to the first numerical value A, the second numerical value B and the third numerical value H, record each remaining historical image as a remaining historical image, annotate the remaining historical images with information, and use all the remaining historical images after annotation to perform preliminary training on the support vector machine model to obtain a preliminary support vector machine model; input a part of the remaining historical images into the preliminary support vector machine model to obtain the predicted annotation information corresponding to the remaining historical images, send the predicted annotation information corresponding to the remaining historical images to the staff so that the staff can correct the predicted annotation information, and input the remaining historical images after the corrected predicted annotation information into the preliminary support vector machine model again to train it to obtain the trained support vector machine model.

3. The prediction method of the poultry growth model based on environmental monitoring according to claim 2, characterized in that: The method of screening all historical images according to the first value A, the second value B and the third value H, recording each remaining historical image as a remaining historical image, and annotating the remaining historical images with information includes: judge Is the value greater than If the value of is greater than, all historical images in this first cluster are deleted; after each first cluster is analyzed, each remaining historical image is recorded as a remaining historical image; all the remaining historical images are clustered again to obtain multiple second clusters; for each second cluster, a preset number of remaining historical images are screened out according to the distance from the center of the second cluster to form a labeled image set, and the labeled image set is sent to the staff for information labeling to obtain the labeling information corresponding to the labeled image set, and the labeling information corresponding to the labeled image set is used as the labeling information corresponding to each of the remaining historical images in the second cluster.

4. The prediction method of the poultry growth model based on environmental monitoring according to claim 1, characterized in that: Each third cluster is compared and analyzed with each second cluster, and each current image is added to the second cluster according to the comparison and analysis result to obtain the second future growth indicator data corresponding to each current image, including: For each third cluster, a feature vector of the cluster center image corresponding to the second cluster is calculated with a feature vector of the cluster center image corresponding to each second cluster to obtain a plurality of second similarities, and when the second similarity is greater than a preset second similarity threshold, the corresponding second cluster is used as a similar cluster of each third cluster; Determine the number of similar clustering clusters. If the number of similar clustering clusters is one, directly define this similar clustering cluster as the most similar clustering cluster corresponding to the third clustering cluster; otherwise, calculate the distance between each third clustering cluster and each corresponding similar clustering cluster, and screen out the most similar clustering cluster corresponding to each third clustering cluster according to the distance calculation result; add the current image in each third clustering cluster to its corresponding most similar clustering cluster, and use the annotation information corresponding to the most similar clustering cluster as the annotation information of the current image added to the most similar clustering cluster, so as to obtain the second future growth indicator data corresponding to each current image.

5. The prediction method of the poultry growth model based on environmental monitoring according to claim 4, characterized in that: Calculate the distance between each third cluster and each corresponding similar cluster, and select the most similar cluster corresponding to each third cluster according to the distance calculation result, including: The distance between each third cluster and each corresponding similar cluster is calculated, wherein a current image is arbitrarily selected from the third cluster, and the distance between the selected current image and each historical image belonging to the similar cluster is calculated, and all distances are averaged and the average is used as the distance between the third cluster and the corresponding similar cluster; the similar cluster corresponding to the smallest distance is used as the most similar cluster corresponding to the third cluster.

6. A prediction system for poultry growth model based on environmental monitoring, characterized in that: include: An acquisition module, used for acquiring historical images and current images of each breeding cage where poultry are raised; A training module, used to perform cluster analysis on all historical images to obtain at least two first clusters, screen all historical images according to the first clusters, cluster each remaining historical image again to obtain multiple second clusters, complete information annotation of each remaining historical image based on the second clusters, and the annotate the information including future growth indicator data; train the support vector machine model using the annotated historical images as training data to obtain a trained support vector machine model; A first calculation module, used for inputting the current image into the trained support vector machine model, and outputting first future growth indicator data corresponding to each current image; The second calculation module is used to perform cluster analysis on all current images to obtain third cluster clusters, compare and analyze each third cluster cluster with each second cluster cluster, add each current image to the second cluster cluster according to the comparison and analysis results, and obtain second future growth indicator data corresponding to each current image; perform weighted summation on the first future growth indicator data and the second future growth indicator data to obtain final future growth indicator data corresponding to each current image.

7. The prediction system of poultry growth model based on environmental monitoring according to claim 6, characterized in that: Training modules include: A first calculation unit is used to analyze each of the first clusters, wherein each of the first clusters contains at least two of the historical images, and count the number of historical image pairs in the first cluster, which is recorded as a first value A, and the historical image pair is composed of any two different historical images; calculate the first similarity of the two historical images in each historical image pair, compare the first similarity with a preset first similarity threshold, and count the number of first similarities that do not exceed the first similarity threshold, which is recorded as a second value B; count the number of historical images contained in the first cluster, which is recorded as a third value H; A training unit is used to complete the screening of all historical images according to the first numerical value A, the second numerical value B and the third numerical value H, record each remaining historical image as a remaining historical image, annotate the remaining historical images with information, and use all the remaining historical images after annotation to perform preliminary training on the support vector machine model to obtain a preliminary support vector machine model; input a part of the remaining historical images into the preliminary support vector machine model to obtain the predicted annotation information corresponding to the remaining historical images, send the predicted annotation information corresponding to the remaining historical images to the staff so that the staff can correct the predicted annotation information, and input the remaining historical images after the corrected predicted annotation information into the preliminary support vector machine model again for training to obtain a trained support vector machine model.

8. The prediction system of poultry growth model based on environmental monitoring according to claim 7, characterized in that: Training modules include: Marking unit, used to judge Is the value greater than If the value of is greater than, all historical images in this first cluster are deleted; after each first cluster is analyzed, each remaining historical image is recorded as a remaining historical image; all the remaining historical images are clustered again to obtain multiple second clusters; for each second cluster, a preset number of remaining historical images are screened out according to the distance from the center of the second cluster to form a labeled image set, and the labeled image set is sent to the staff for information labeling to obtain the labeling information corresponding to the labeled image set, and the labeling information corresponding to the labeled image set is used as the labeling information corresponding to each of the remaining historical images in the second cluster.

9. The prediction system of poultry growth model based on environmental monitoring according to claim 6, characterized in that: The second computing module includes: a second calculation unit, configured to calculate, for each third cluster, a similarity between a feature vector of a cluster center image corresponding to the second cluster and a feature vector of a cluster center image corresponding to each second cluster, to obtain a plurality of second similarities, and when the second similarity is greater than a preset second similarity threshold, use the corresponding second cluster as a similar cluster of each third cluster; The third calculation unit is used to determine the number of similar clusters. If the number of similar clusters is one, this similar cluster is directly defined as the most similar cluster corresponding to the third cluster; otherwise, the distance between each third cluster and each corresponding similar cluster is calculated, and the most similar cluster corresponding to each third cluster is screened out according to the distance calculation result; the current image in each third cluster is added to the corresponding most similar cluster, and the annotation information corresponding to the most similar cluster is used as the annotation information of the current image added to the most similar cluster, so as to obtain the second future growth indicator data corresponding to each current image.

10. The prediction system of poultry growth model based on environmental monitoring according to claim 9, characterized in that: The third computing unit includes: The fourth calculation unit is used to calculate the distance between each third cluster and each corresponding similar cluster, wherein a current image is arbitrarily selected from the third cluster, and the distance between the selected current image and each historical image belonging to the similar cluster is calculated, all distances are averaged, and the average is used as the distance between the third cluster and the corresponding similar cluster; the similar cluster corresponding to the smallest distance is used as the most similar cluster corresponding to the third cluster.

Images