Unlock instant, AI-driven research and patent intelligence for your innovation.

Pig state analysis system and method based on visual AI

A state analysis and visual technology, applied in character and pattern recognition, instruments, and other equipment, can solve problems such as raising breeding costs, wasting labor, and difficult management, achieving good operating speed and recognition accuracy, and reducing manual monitoring. Error, the effect of reducing the risk of infection

Pending Publication Date: 2021-08-17
CHONGQING ACAD OF ANIMAL SCI +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] General pig breeding methods include indoor captive breeding and outdoor free-range breeding, mainly relying on human breeding, too much human intervention, not only can not avoid bacterial infection, resulting in low meat yield, wasting a lot of labor, increasing the cost of breeding, and difficult It is impossible to provide accurate reference for the operation of pig farms by monitoring the activity status of pigs all the time. In large-scale intensive livestock farms, manual on-site care is still required, and problems such as group infection of pigs cannot be avoided. There are management difficulties in the farm

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Pig state analysis system and method based on visual AI
  • Pig state analysis system and method based on visual AI

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] See attached figure 1 , 2 , in this embodiment, the pose state of the pig includes two states of standing and lying down, and other subdivided states (such as standing A, standing B) can be realized by defining edge information and behavior pose information similarly.

[0039] The source video acquisition terminal includes an optical sensing module and a distortion processing module. The optical sensing module sequentially collects images of the pigsty site with a resolution of 1280*720, and each frame of data collected is transmitted to the distortion processing module for distortion processing. The distortion processing module uses the linear distortion correction algorithm of the Go board image to de-distort each input frame, and then encodes it into a real-time video stream in h264 format at a frame rate of 25FPS, and transmits it to the image database for storage.

[0040] For the preliminary training process of the discriminant model, the data indexing module rec...

Embodiment 2

[0047] Such as figure 1 , 2 As shown, this embodiment discloses a method for analyzing the state of pigs based on visual AI, including:

[0048] Obtain the pig image in the video stream format obtained by image acquisition of the pigsty. The reason why the video stream format is required is to track and analyze the pig image frame by frame, so as to facilitate real-time understanding of the change of the pig's pose. The acquisition process of the pig image in the video stream format includes: collecting the image data of the pigsty site, performing distortion correction on the collected data, and obtaining the pig image; combining the obtained pig image frame by frame in the order of the collected data video stream. Among them, the image data of the pigsty site is sequentially collected by the optical sensing module with a resolution of 1280*720. The distortion correction process uses the Go board image linear distortion correction algorithm to de-distort each frame of imag...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a pig state analysis system and method based on visual AI. The method comprises the following steps: acquiring a pig image in a video stream format obtained by performing image acquisition on a pigsty; respectively indexing the edge information and the behavior attitude information of each acquired pig image, and training a classifier model; analyzing a to-be-discriminated current frame of pig image by using the trained discrimination model, and discriminating edge information and behavior attitude information of the current frame of pig image; recording the edge information of the current frame of pig image, evaluating the similarity with the edge information of each pig in the previous frame, calculating the Euclidean distance between the edge information of each pig in the previous frame and the mass center of the edge information of each pig in the previous frame, carrying out weighted summation to obtain a matching score, completing matching of one pig at the maximum score, and circulating the process to complete matching of all pigs, and counting the matching pairs as the total number of pigs. According to the method, the total number of pigs can be remotely counted in real time, the posture state of the pigs is analyzed, instance individuals do not need to be recognized, and the better running speed and recognition accuracy are achieved.

Description

technical field [0001] The invention relates to the field of animal husbandry, in particular to a visual AI-based pig state analysis system and method. Background technique [0002] General pig breeding methods include indoor captive breeding and outdoor free-range breeding, mainly relying on human breeding, too much human intervention, not only can not avoid bacterial infection, resulting in low meat yield, wasting a lot of labor, increasing the cost of breeding, and difficult It is impossible to provide accurate reference for the operation of pig farms by monitoring the activity status of pigs all the time. In large-scale intensive livestock farms, manual on-site care is still required, and problems such as group infection of pigs cannot be avoided. Management difficulties exist in farms. Contents of the invention [0003] The purpose of the present invention is to provide a pig state analysis system and method based on visual AI to analyze and count the pose of pigs ba...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/62A01K29/00
CPCA01K29/005G06V40/10G06F18/22G06F18/241Y02P90/30
Inventor 龙定彪王浩杨飞云徐顺来蒲施桦曾雅琼朱佳明简悦朱萌钢
Owner CHONGQING ACAD OF ANIMAL SCI