Feeding control and display method and system for pig farm

Abstract

The invention relates to the technical field of intelligent management of pig farms, and particularly discloses a feeding control and display method and system for a pig farm, and the method comprises the steps: obtaining a pig house image containing a timestamp based on a camera installed in a pig house, carrying out the recognition of the pig house image containing the timestamp, and obtaining the outlines of pigs at different moments; identifying the outline of the pig, judging the weight of the pig, obtaining the position of the pig, and determining the feeding probability according to the weight of the pig and the position of the pig; selecting a feeding point position in the pig house based on the feeding probability, and regularly generating a feeding instruction pointing to the feeding point position; according to the system, the conditions of the pigs are obtained through the camera, the feeding probability of each pig is determined, and then different feeding points are selected, so that the feeding probabilities of the pigs are finally similar, the uniformity of the growth process of the pigs is greatly improved, and the intelligent level is extremely high.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management of pig farms, and specifically to a feeding control and display method and system for a pig farm. Background Art

[0002] The feeding control of a pig farm is a key link in modern breeding management, which is related to the growth rate, feed utilization rate, health status and economic benefits of pigs. Most of the existing intelligent feeding control schemes for pig farms are simple fixed-time and fixed-point feeding schemes, without considering the actual situation of the pigsty, and it is easy to produce a large difference in growth in the pigsty. How to provide a more intelligent feeding scheme is the technical problem that the technical solution of the present invention wants to solve. Summary of the Invention

[0003] An object of the present invention is to provide a feeding control and display method and system for a pig farm to solve the problems raised in the above background art.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A feeding control and display method and system for a pig farm, the method comprising:

[0006] Obtaining a pigsty image with a time stamp based on a camera installed in the pigsty, and recognizing the pigsty image with the time stamp to obtain pig silhouettes at different times;

[0007] Recognizing the pig silhouettes, determining the weight of the pigs, obtaining the positions of the pigs, and determining the feeding probability according to the weight and position of the pigs;

[0008] Selecting a feeding point in the pigsty based on the feeding probability, and regularly generating a feeding instruction pointing to the feeding point;

[0009] Constructing a pig farm model according to the positional relationship of all pigsties, locating abnormal pigs according to the pig silhouettes at each moment in each pigsty, determining a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model;

[0010] Synchronously adjusting the image acquisition frequency of the cameras in the pigsty according to the generation time of the feeding instruction and the warning layer.

[0011] As a further aspect of the present invention: The step of recognizing the pig silhouettes, determining the weight of the pigs, obtaining the positions of the pigs, and determining the feeding probability according to the weight and position of the pigs includes:

[0012] Recognizing the pig silhouettes, and locating the head region and the tail region;

[0013] Obtain the center point of the tail area, and obtain the contour point on the contour of the head area that is farthest from the center point of the tail area as the pig position;

[0014] Input the pig contour into the trained weight evaluation model to obtain the pig weight; the weight evaluation model is a neural network model, with the input being the pig contour and the output being the pig weight;

[0015] Determine the feeding probability according to the pig weight and the pig position;

[0016] The process of determining the feeding probability is as follows:

[0017] S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, S i represents the feeding value, which is an intermediate parameter, α, β, and γ are preset correction coefficients, W is the pig weight, D1 is the distance between the i-th pig and the pig with the largest weight, and D2 is the distance between the i-th pig and the door of the pigsty.

[0018] As a further solution of the present invention: the step of selecting a feeding point in the pigsty based on the feeding probability and periodically generating a feeding instruction pointing to the feeding point includes:

[0019] Query all the feeding points in the pigsty;

[0020] Adjust the feeding probability based on the distance between the feeding point and each pig; the adjustment process is: where P ′ is the adjusted feeding probability, P is the feeding probability before adjustment; D3 is the distance between the feeding point and the pig;

[0021] Calculate the standard deviation of the corrected feeding probability, and select the feeding point with the smallest standard deviation;

[0022] Periodically generate a feeding instruction pointing to the feeding point; where the single feeding amount is a preset fixed value.

[0023] As a further solution of the present invention: the step of constructing a pig farm model according to the positional relationship of all pigsties, positioning abnormal pigs according to the pig contours at each moment in each pigsty, determining a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model includes:

[0024] Obtain the relative position of the pigsty relative to the origin of the pig farm, and construct a pig farm model based on the relative position by statistically analyzing the building information of the pigsty and the pig contour at the latest moment; the scale of the pig farm model is determined by the size of the pig farm and the display size;

[0025] For any pigsty, locate the pigs in the pig silhouette at the latest moment whose weights are less than a preset weight threshold, and extract pig features from the pig silhouette; wherein, when the weight of a pig reaches the preset weight threshold, the extracted pig features are deleted;

[0026] Track the corresponding pigs in the images at each moment based on the pig features to obtain the food intake; the calculation process of the food intake is: the feeding speed multiplied by the feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs;

[0027] Compare the food intake with a preset food intake threshold, generate a warning layer, and overlay it on the area corresponding to the pigsty in the pig farm model.

[0028] As a further solution of the present invention: the step of synchronously adjusting the image acquisition frequency of the camera in the pigsty according to the generation time of the feeding instruction and the warning layer includes:

[0029] Calculate the time difference between the current moment and the generation time of the nearest feeding instruction;

[0030] For any pigsty, query the corresponding area in the pig farm model, synchronously query the warning layer of this area, and determine the feature value according to the warning layer;

[0031] Adjust the image acquisition frequency of the camera in this pigsty according to the time difference and the feature value.

[0032] As a further solution of the present invention: the method further includes:

[0033] Real-time obtain audio information based on the microphone installed in the pigsty, identify the audio information based on a preset audio feature library, and obtain the pigsty requirements; the audio feature library includes audio feature items and requirement items;

[0034] When the pigsty requirements reach the preset requirement conditions, generate a random feeding instruction;

[0035] Synchronously generate a real-time acquisition instruction pointing to the camera, identify the pigsty image collected in real time, and correct the feeding probability of the pigs.

[0036] The technical solution of the present invention also provides a feeding control and display system for a pig farm, and the system includes:

[0037] A pig image acquisition module, which is used to obtain pigsty images with time stamps based on the cameras installed in the pigsty, and identify the pigsty images with time stamps to obtain pig silhouettes at different moments;

[0038] A pig image recognition module, which is used to recognize the pig silhouette, determine the weight of the pig, obtain the position of the pig, and determine the feeding probability according to the weight and position of the pig;

[0039] A feeding instruction generation module, configured to select a feeding point in a pigsty based on the feeding probability and generate a feeding instruction pointing to the feeding point at regular intervals;

[0040] A model establishment and display module, configured to construct a pig farm model according to the positional relationship of all pigsties, locate abnormal pigs according to the pig outlines at each moment in each pigsty, determine a warning layer, and superimpose it on the area corresponding to the pigsty in the pig farm model;

[0041] An acquisition frequency update module, configured to synchronously adjust the image acquisition frequency of the cameras in the pigsty according to the generation moment of the feeding instruction and the warning layer.

[0042] As a further solution of the present invention: the pig image recognition module includes:

[0043] A contour recognition unit, configured to recognize the pig contour and locate the head area and the tail area;

[0044] A position extraction unit, configured to obtain the center point of the tail area, and obtain the contour point on the contour of the head area that is farthest from the center point of the tail area as the pig position;

[0045] A weight recognition unit, configured to input the pig contour into a trained weight evaluation model to obtain the pig weight; the weight evaluation model is a neural network model, with the pig contour as the input and the pig weight as the output;

[0046] A probability determination unit, configured to determine the feeding probability according to the pig weight and the pig position;

[0047] The process of determining the feeding probability is as follows:

[0048] S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, S i represents the feeding value, which is an intermediate parameter, α, β, and γ are preset correction coefficients, W is the pig weight, D1 is the distance between the i-th pig and the pig with the largest pig weight, and D2 is the distance between the i-th pig and the door of the pigsty.

[0049] As a further solution of the present invention: the feeding instruction generation module includes:

[0050] A point position query unit, configured to query all feeding points in the pigsty;

[0051] A probability test unit, configured to adjust the feeding probability based on the distance between the feeding point and each pig; the adjustment process is: where P ′Let \(P_{adj}\) be the adjusted feeding probability and \(P\) be the feeding probability before adjustment; \(D_3\) is the distance between the feeding point and the pigs.

[0052] A point selection unit, which is used to calculate the standard deviation of the corrected feeding probability and select the feeding point with the smallest standard deviation.

[0053] An execution unit, which is used to generate a feeding instruction pointing to the feeding point at regular intervals; where the single feeding amount is a preset fixed value.

[0054] As a further solution of the present invention: the model establishment and display module includes:

[0055] A pig farm model construction unit, which is used to obtain the relative position of the pigsty relative to the origin of the pig farm, and construct a pig farm model based on the relative position by counting the building information of the pigsty and the pig silhouette at the latest moment; the scale of the pig farm model is determined by the size of the pig farm and the display size.

[0056] A pig feature extraction unit, which is used for any pigsty to locate the pigs with a weight less than a preset weight threshold in the pig silhouette at the latest moment, and extract pig features from the pig silhouette; where when the weight of the pig reaches the preset weight threshold, the extracted pig features are deleted.

[0057] A pig tracking unit, which is used to track the corresponding pigs in the images at each moment based on the pig features and obtain the feeding amount; the calculation process of the feeding amount is: feeding speed multiplied by feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs.

[0058] A layer generation and superposition unit, which is used to compare the feeding amount with a preset feeding amount threshold, generate a warning layer, and superpose it on the area corresponding to the pigsty in the pig farm model.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] The present invention obtains the situation of pigs through a camera, determines the feeding probability of each pig, and then selects different feeding points, so that the feeding probabilities of each pig are finally similar, greatly improving the uniformity in the growth process of pigs and having a very high level of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0062] Figure 1 Shows the overall flow block diagram of the feeding control and display method for a pig farm.

[0063] Figure 2The structure diagram of the feeding control and display system for a pig farm is shown. Detailed implementation mode

[0064] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0065] Figure 1 It is the overall flowchart of the feeding control and display method and system for a pig farm. In an embodiment of the present invention, a feeding control and display method for a pig farm, the method includes:

[0066] Step S100: Obtain a pigsty image with a timestamp based on a camera installed in the pigsty, and identify the pig silhouettes at different times from the pigsty image with the timestamp.

[0067] This application is applied to a pig farm. There are multiple pig-raising units in the pig farm, called pigsties. Cameras are installed in the pigsties. Based on the cameras installed in the pigsties, pigsty images are obtained. When obtaining the images, the acquisition time, called the timestamp, needs to be recorded; by identifying the pigsty images at each time, the pig silhouettes can be located. Generally speaking, the pig silhouette is the silhouette of each pig in the image.

[0068] Step S200: Identify the pig silhouettes, determine the pig weight, obtain the pig position, and determine the feeding probability according to the pig weight and the pig position.

[0069] The pig silhouette actually reflects the pig area, corresponding to a certain pig. By identifying the pig silhouette, the pig weight and the pig position can be obtained. According to the pig weight and the pig position, the feeding probability of the pig can be determined; generally, the greater the pig weight, the higher the feeding probability, and the closer the pig position is to the feeding port, the higher the feeding probability.

[0070] Step S300: Select a feeding point in the pigsty based on the feeding probability, and regularly generate a feeding instruction pointing to the feeding point.

[0071] Select the feeding point according to the feeding probability of each pig. Every time a preset period passes, a feeding is carried out. When feeding, generate a feeding instruction pointing to the feeding point and control it to feed; regarding the hardware of the feeding point, generally it is a valve, and the feeding instruction is the timed opening instruction of the valve. For how long to open, for more complex equipment, the feeding amount can be controlled. There are many feeding hardwares in the prior art and will not be elaborated here.

[0072] Step S400: Construct a pig farm model based on the positional relationships of all pig houses, locate abnormal pigs according to the pig outlines at each moment in each pig house, determine a warning layer, and overlay it on the area corresponding to the pig house in the pig farm model;

[0073] Steps S100 to S300 occur simultaneously in each pig house. Construct a pig farm model based on the positional relationships of all pig houses. The pig farm model is a three-dimensional model for display. Locate abnormal pigs according to the pig outlines at each moment in each pig house, create a warning layer simultaneously according to the positioning results, and overlay it on the area corresponding to the pig house in the pig farm model; The warning layer is generally a layer of a certain hue, such as a red layer, and the transparency is adjusted according to the abnormality of the abnormal pigs. The higher the degree of abnormality, the higher the transparency, which is used to prompt the management personnel to manage the corresponding pig house.

[0074] Step S500: Synchronously adjust the image acquisition frequency of the cameras in the pig houses according to the generation moment of the feeding instruction and the warning layer;

[0075] Each time a feeding instruction is generated, the pigs will be temporarily stable for a period of time. Therefore, the image acquisition frequency of the cameras can be optimized according to the generation moment of the feeding instruction. Each time a feeding instruction is generated, the image acquisition frequency is set to a low value and gradually increases over time; At the same time, according to the generation situation of the warning layer, the image acquisition frequency of the cameras can also be adjusted. The higher the degree of abnormality, the higher the image acquisition frequency.

[0076] Regarding step S200, the steps of identifying the pig outline, determining the pig weight, obtaining the pig position, and determining the feeding probability according to the pig weight and the pig position include:

[0077] Identify the pig outline and locate the head region and the tail region;

[0078] Obtain the center point of the tail region, and obtain the contour point on the outline of the head region that is farthest from the center point of the tail region as the pig position;

[0079] Input the pig outline into a trained weight evaluation model to obtain the pig weight; The weight evaluation model is a neural network model, with the pig outline as the input and the pig weight as the output;

[0080] Determine the feeding probability according to the pig weight and the pig position;

[0081] In an example of the technical solution of the present invention, the process of determining the feeding probability is described. The contour of the pig is recognized, the head region and the tail region are located, the center point of the tail region is obtained, and the contour point on the contour of the head region that is farthest from the center point of the tail region is used as the position of the pig. This process is actually to try to determine the position of the pig's mouth in the pig's contour. Since the pig is constantly moving and its posture often changes, it is difficult to directly locate the pig's mouth. By this method, the position closest to the pig's mouth can be determined as a whole. Of course, if a recognition model with a higher cost is used, a more accurate position of the pig's mouth can also be located, but the cost is too high and it is a bit uneconomical. Based on the conventional positioning scheme, this application has made some fine-tuning and determined a relatively accurate position with high cost performance. In other words, since the pig's head and pig's tail have obvious characteristics and their positions are also at both ends of the pig, the positioning process is extremely easy.

[0082] The contour of the pig is input into the trained weight evaluation model to obtain the weight of the pig. The weight evaluation model is a neural network model, and its training process is as follows: First, the staff selects pigs with different weights and takes a large number of images at different angles in their different postures. The contour is extracted from the images, and the contour and the weight are used as a sample. When there are enough samples, the trained neural network model is the weight evaluation model.

[0083] After obtaining the weight of the pig and the position of the pig, the feeding probability can be determined according to the weight of the pig and the position of the pig. The process of determining the feeding probability is as follows:

[0084] S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, S i represents the feeding value, which is an intermediate parameter, α, β, and γ are preset correction coefficients, W is the weight of the pig, D1 is the distance between the i-th pig and the pig with the largest weight, and D2 is the distance between the i-th pig and the door of the pigsty.

[0085] In the above content, the symbols of α, β, and γ are generally not limited. β mostly takes negative values. Of course, it can also take zero. The key point is that the influencing parameters of the feeding probability are limited to the weight of the pig, the distance from the pig with the largest weight, and the distance from the door of the pigsty. The distance from the door of the pigsty affects the possibility of being noticed during the inspection by the management staff.

[0086] Regarding step S300, the step of selecting a feeding point in the pigsty based on the feeding probability and generating a feeding instruction pointing to the feeding point at regular intervals includes:

[0087] Query all the feeding points in the pigsty;

[0088] Adjust the feeding probability based on the distance between the feeding point and each pig; the adjustment process is as follows: In the formula, P ′ is the adjusted feeding probability, and P is the feeding probability before adjustment; D3 is the distance between the feeding point and the pig;

[0089] Calculate the standard deviation of the corrected feeding probability, and select the feeding point with the smallest standard deviation;

[0090] Generate a feeding instruction pointing to the feeding point at regular intervals; among them, the single feeding amount is a preset fixed value.

[0091] Query all feeding points in the pigsty. The feeding points are determined according to the feeding equipment, and the feeding port is the feeding point. For the technical solution of the present invention, the feeding point is a known position. Analyze each feeding point and select the final feeding point.

[0092] Specifically, the process of analyzing each feeding point and selecting the final feeding point is as follows: adjust the feeding probability based on the distance between the feeding point and each pig, calculate the standard deviation of the corrected feeding probability, and select the feeding point with the smallest standard deviation as the final feeding point. This means that the feeding points selected in this application need to ensure that the feeding probabilities of each pig are similar.

[0093] Regarding step S400, the steps of constructing a pig farm model according to the positional relationship of all pigsties, locating abnormal pigs according to the pig profiles at each moment in each pigsty, determining a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model include:

[0094] Obtain the relative position of the pigsty relative to the origin of the pig farm, and construct a pig farm model based on the building information of the pigsty and the pig profiles at the latest moment based on the relative position; the scale of the pig farm model is determined by the size of the pig farm and the display size;

[0095] For any pigsty, locate the pigs with a weight less than the preset weight threshold in the pig profile at the latest moment, and extract pig features in the pig profile; among them, when the pig weight reaches the preset weight threshold, the extracted pig features are deleted;

[0096] Track the corresponding pigs in the images at each moment based on the pig features, and obtain the feeding amount; the calculation process of the feeding amount is: feeding speed multiplied by feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs;

[0097] Compare the feeding amount with the preset feeding amount threshold, generate a warning layer, and superimpose it on the area corresponding to the pigsty in the pig farm model.

[0098] The pig farm model is a three-dimensional model used to inform the management staff of the situation of the pig farm, obtain the relative position of the pigsty with respect to the origin of the pig farm, and construct the pig farm model based on the relative position by counting the building information of the pigsty and the pig silhouette at the latest moment. The scale of the constructed pig farm model is determined by the size of the pig farm and the display size; for any pigsty, pigs with a weight less than a preset weight threshold are located in the pig silhouette at the latest moment as abnormal pigs, and pig features are extracted from the pig silhouette. The extraction process can use conventional image feature extraction methods, and this extraction is carried out over time and will be updated over time. The located abnormal pigs may not be unique. The pig features are stored, and when the pig weight reaches the preset weight threshold, the extracted pig features are deleted.

[0099] Regarding step S500, the step of synchronously adjusting the image acquisition frequency of the camera in the pigsty according to the generation time of the feeding instruction and the warning layer includes:

[0100] Calculate the time difference between the current time and the generation time of the nearest feeding instruction;

[0101] For any pigsty, query the corresponding area in the pig farm model, synchronously query the warning layer of this area, and determine the characteristic value according to the warning layer;

[0102] Adjust the image acquisition frequency of the camera in this pigsty according to the time difference and the characteristic value.

[0103] Query the generation time of the nearest feeding instruction, calculate the time difference between the current time and the generation time. For any pigsty, query the corresponding area in the pig farm model, synchronously query the warning layer of this area, and determine the characteristic value. The characteristic value can determine a base value by hue and then divide it by the transparency. The smaller the transparency, the more the warning layer receives the corresponding hue, and the greater the risk value; finally, the image acquisition frequency can be determined according to the time difference and the characteristic value.

[0104] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0105] Based on the microphone installed in the pigsty, real-time audio information is obtained, and the audio information is recognized based on a preset audio feature library to obtain the pigsty requirements; the audio feature library includes audio feature items and requirement items;

[0106] When the pigsty requirements reach the preset requirement conditions, a random feeding instruction is generated;

[0107] Synchronously generate a real-time acquisition instruction pointing to the camera, and recognize the pigsty image collected in real time to correct the feeding probability of the pigs.

[0108] The camera generally has a built-in microphone, which can also be called an audio detector. It obtains audio information in real time based on the microphone installed in the pigsty, and identifies the audio information based on a preset audio feature library to obtain the pigsty requirements. The pigsty requirements represent the degree of food demand of each pig in the pigsty. When the pigsty requirements reach the preset demand conditions, a random feeding instruction is generated. The random feeding instruction randomly selects a feeding point and generates a feeding instruction. This process is an auxiliary solution for supplementing food for pigs and will only be triggered when there are obvious corresponding audio features, generally occurring when the feeding device is damaged. For example, when it is time to feed but no feeding occurs, after a period of time in the pigsty, there will be sounds, and then obvious audio features will be detected.

[0109] In addition, during the random feeding process, the activities of the pigs are relatively intense. At this time, a high-frequency acquisition instruction pointing to the camera is generated, that is, a real-time acquisition instruction, to identify the pigsty images collected in real time and correct the feeding probability of the pigs. Specifically, the way to correct the feeding probability of the pigs can be to correct the recognition results of the pig contours, such as the pig weight and pig position. When the pigs are moving, the shooting postures and angles are more diverse, and the recognition results are more accurate.

[0110] Figure 2 The structure diagram of the feeding control and display system of the pig farm is shown. In a preferred embodiment of the technical solution of the present invention, a feeding control and display system of the pig farm is also provided. The system 10 includes:

[0111] A pig image acquisition module 11, which is used to acquire pigsty images with time stamps based on the cameras installed in the pigsty, and identify the pigsty images with time stamps to obtain pig contours at different times.

[0112] A pig image recognition module 12, which is used to identify the pig contours, determine the pig weight, obtain the pig position, and determine the feeding probability according to the pig weight and pig position.

[0113] A feeding instruction generation module 13, which is used to select a feeding point in the pigsty based on the feeding probability and generate a feeding instruction pointing to the feeding point at regular intervals.

[0114] A model establishment and display module 14, which is used to construct a pig farm model according to the positional relationship of all pigsties, locate abnormal pigs according to the pig contours at each moment in each pigsty, determine a warning layer, and superimpose it on the area corresponding to the pigsty in the pig farm model.

[0115] An acquisition frequency update module 15, which is used to synchronously adjust the image acquisition frequency of the cameras in the pigsty according to the generation time of the feeding instruction and the warning layer.

[0116] Furthermore, the pig image recognition module 12 includes:

[0117] A contour recognition unit for recognizing the contour of a pig and locating the head region and the tail region;

[0118] A position extraction unit for obtaining the center point of the tail region and obtaining the contour point on the contour of the head region that is farthest from the center point of the tail region as the position of the pig;

[0119] A weight recognition unit for inputting the pig contour into a trained weight evaluation model to obtain the pig weight; the weight evaluation model is a neural network model, with the input being the pig contour and the output being the pig weight;

[0120] A probability determination unit for determining the feeding probability according to the pig weight and the pig position;

[0121] The process of determining the feeding probability is as follows:

[0122] S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, S i represents the feeding value, which is an intermediate parameter, α, β, and γ are preset correction coefficients, W is the pig weight, D1 is the distance between the i-th pig and the pig with the largest weight, and D2 is the distance between the i-th pig and the door of the pigsty.

[0123] Specifically, the feeding instruction generation module 13 includes:

[0124] A point position query unit for querying all feeding point positions in the pigsty;

[0125] A probability test unit for adjusting the feeding probability based on the distance between the feeding point position and each pig; the adjustment process is: where P' is the adjusted feeding probability, P is the feeding probability before adjustment; D3 is the distance between the feeding point position and the pig;

[0126] A point position selection unit for calculating the standard deviation of the corrected feeding probability and selecting the feeding point position with the smallest standard deviation;

[0127] An execution unit for periodically generating a feeding instruction pointing to the feeding point position; where the single feeding amount is a preset fixed value.

[0128] Furthermore, the model establishment and display module 14 includes:

[0129] A pig farm model construction unit for obtaining the relative position of the pigsty with respect to the origin of the pig farm and constructing a pig farm model based on the relative position by statistically analyzing the building information of the pigsty and the pig contour at the latest moment; the scale of the pig farm model is determined by the size of the pig farm and the display size;

[0130] The pig feature extraction unit is used to locate pigs with a weight less than a preset weight threshold in the pig contour at the latest moment in any pigsty, and extract pig features from the pig contour; wherein, when the pig weight reaches the preset weight threshold, the extracted pig features are deleted;

[0131] The pig tracking unit is used to track the corresponding pigs in the images at each moment based on the pig features and obtain the feeding amount; the calculation process of the feeding amount is: feeding speed multiplied by feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs;

[0132] The layer generation and overlay unit is used to compare the feeding amount with a preset feeding amount threshold, generate a warning layer, and overlay it on the area corresponding to the pigsty in the pig farm model.

[0133] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A feeding control and display method for a pig farm, characterized in that, The method includes: Obtaining a pigsty image with a timestamp based on a camera installed in the pigsty, and identifying the pigsty image with the timestamp to obtain pig silhouettes at different times; Identifying the pig silhouettes, determining the weight of the pigs, obtaining the positions of the pigs, and determining the feeding probability according to the weight and position of the pigs; Selecting a feeding point in the pigsty based on the feeding probability, and periodically generating a feeding instruction pointing to the feeding point; Constructing a pig farm model according to the positional relationship of all pigsties, locating abnormal pigs based on the pig silhouettes at each moment in each pigsty, determining a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model; Synchronously adjusting the image acquisition frequency of the cameras in the pigsty according to the generation time of the feeding instruction and the warning layer.

2. The feeding control and display method of the pig farm according to claim 1, characterized in that The step of identifying the pig silhouettes, determining the weight of the pigs, obtaining the positions of the pigs, and determining the feeding probability according to the weight and position of the pigs includes: Identifying the pig silhouettes and locating the head region and the tail region; Obtaining the center point of the tail region, and obtaining the contour point on the contour of the head region that is farthest from the center point of the tail region as the position of the pig; Inputting the pig silhouette into a trained weight evaluation model to obtain the weight of the pig; the weight evaluation model is a neural network model, with the input being the pig silhouette and the output being the weight of the pig; Determining the feeding probability according to the weight and position of the pig; The process of determining the feeding probability is: S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, and S i represents the feeding value, which is an intermediate parameter. α, β, and γ are preset correction coefficients, W is the weight of the pig, D1 is the distance between the i-th pig and the pig with the largest weight, and D2 is the distance between the i-th pig and the door of the pigsty.

3. The feeding control and display method of the pig farm according to claim 2, characterized in that, The step of selecting a feeding point in the pigsty based on the feeding probability and periodically generating a feeding instruction pointing to the feeding point includes: Querying all feeding points in the pigsty; Adjust the feeding probability based on the distance between the feeding point and each pig; the adjustment process is as follows: In the formula, P ′ is the adjusted feeding probability, P is the feeding probability before adjustment; D3 is the distance between the feeding point and the pig Calculating the standard deviation of the corrected feeding probability, and selecting the feeding point with the smallest standard deviation; Periodically generating a feeding instruction pointing to the feeding point; wherein, the single feeding amount is a preset fixed value.

4. The feeding control and display method of the pig farm according to claim 3, characterized in that, The step of constructing a pig farm model according to the positional relationship of all pigsties, locating abnormal pigs based on the pig silhouettes at each moment in each pigsty, determining a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model includes: Obtaining the relative position of the pigsty relative to the origin of the pig farm, and constructing a pig farm model based on the relative position by statistically analyzing the building information of the pigsty and the pig silhouettes at the latest moment; the scale of the pig farm model is determined by the size of the pig farm and the display size; For any pigsty, locating pigs with a weight less than a preset weight threshold in the pig silhouette at the latest moment, and extracting pig features from the pig silhouette; wherein, when the weight of the pig reaches the preset weight threshold, the extracted pig features are deleted; Tracking the corresponding pigs in the images at each moment based on the pig features to obtain the feeding amount; the calculation process of the feeding amount is: feeding speed multiplied by feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs; Comparing the feeding amount with a preset feeding amount threshold, generating a warning layer, and superimposing it on the area corresponding to the pigsty in the pig farm model.

5. The feeding control and display method for a pig farm according to claim 1, wherein, The step of synchronously adjusting the image acquisition frequency of the cameras in the pigsty according to the generation time of the feeding instruction and the warning layer includes: Calculating the time difference between the current moment and the generation time of the nearest feeding instruction; For any pigsty, querying the corresponding area in the pig farm model, synchronously querying the warning layer of the area, and determining the characteristic value according to the warning layer; Adjust the image acquisition frequency of the camera in the pigsty according to the time difference and the eigenvalue.

6. The feeding control and display method for a pig farm according to claim 1, characterized in that The method further includes: Real-time obtain audio information based on a microphone installed in the pigsty, and identify the audio information based on a preset audio feature library to obtain the pigsty requirements; the audio feature library includes audio feature items and requirement items; When the pigsty requirements reach the preset requirement conditions, generate a random feeding instruction; Synchronously generate a real-time acquisition instruction pointing to the camera, identify the real-time acquired pigsty images, and correct the feeding probability of the pigs.

7. A feeding control and display system for a pig farm, characterized in that, The system includes: A pig image acquisition module, configured to obtain pigsty images with timestamps based on a camera installed in the pigsty, and identify the pigsty images with timestamps to obtain pig silhouettes at different times; A pig image recognition module, configured to recognize the pig silhouettes, determine the pig weight, obtain the pig position, and determine the feeding probability according to the pig weight and the pig position; A feeding instruction generation module, configured to select a feeding point in the pigsty based on the feeding probability and regularly generate a feeding instruction pointing to the feeding point; A model establishment and display module, configured to construct a pig farm model according to the positional relationship of all pigsties, locate abnormal pigs according to the pig silhouettes at each moment in each pigsty, determine a warning layer, and superimpose it on the area corresponding to the pigsty in the pig farm model; An acquisition frequency update module, configured to synchronously adjust the image acquisition frequency of the camera in the pigsty according to the generation time of the feeding instruction and the warning layer.

8. The feeding control and display system for a pig farm according to claim 7, characterized in that, The pig image recognition module includes: A contour recognition unit, configured to recognize the pig silhouettes and locate the head area and the tail area; A position extraction unit, configured to obtain the center point of the tail area, and obtain the contour point on the contour of the head area that is farthest from the center point of the tail area as the pig position; A weight recognition unit, configured to input the pig silhouette into a trained weight evaluation model to obtain the pig weight; the weight evaluation model is a neural network model, with the input being the pig silhouette and the output being the pig weight; A probability determination unit, configured to determine the feeding probability according to the pig weight and the pig position; The process of determining the feeding probability is: S i = αW + βD1 + γD2; where P i is the feeding probability of the i-th pig, S i represents the feeding value, which is an intermediate parameter, α, β, and γ are preset correction coefficients, W is the weight of the pig, D1 is the distance between the i-th pig and the pig with the largest weight, and D2 is the distance between the i-th pig and the door of the pigsty.

9. The feeding control and display system for a pig farm according to claim 8, characterized in that, The feeding instruction generation module includes: A point location query unit, configured to query all feeding points in the pigsty; A probability test unit is used to adjust the feeding probability based on the distance between the feeding point and each pig; the adjustment process is as follows: In the formula, P ′ is the adjusted feeding probability, P is the feeding probability before adjustment; D3 is the distance between the feeding point and the pig. A point location selection unit, configured to calculate the standard deviation of the corrected feeding probability and select the feeding point with the smallest standard deviation; An execution unit, configured to regularly generate a feeding instruction pointing to the feeding point; wherein, the single feeding amount is a preset fixed value.

10. The feeding control and display system for a pig farm according to claim 9, characterized in that, The model establishment and display module includes: A pig farm model construction unit, configured to obtain the relative position of the pigsty with respect to the origin of the pig farm, and construct a pig farm model based on the relative position by counting the building information of the pigsty and the pig silhouettes at the latest moment; the scale of the pig farm model is determined by the size of the pig farm and the display size; A pig feature extraction unit, configured to, for any pigsty, locate pigs with a pig weight less than a preset weight threshold in the pig silhouettes at the latest moment, and extract pig features from the pig silhouettes; wherein, when the pig weight reaches the preset weight threshold, the extracted pig features are deleted; A pig tracking unit, which is used to track corresponding pigs in images at each moment based on pig characteristics and obtain the food intake; the calculation process of the food intake is: the feeding speed multiplied by the feeding duration, and the feeding speed is inversely proportional to the number of adjacent pigs. A layer generation and overlay unit, which is used to compare the food intake with a preset food intake threshold, generate a warning layer, and overlay it on the area corresponding to the pigsty in the pig farm model.

Images