System and method for recognizing posture of single pig body based on stacked hourglass network

A recognition method and recognition system technology, applied in the field of single pig body gesture recognition system, can solve problems such as large manpower and wrong judgment.

Pending Publication Date: 2020-12-18
CHINA AGRI UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method consumes a lot of manpower and because people cannot maintain concentration for a long time, this will inevitably cause certain misjudgments

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
  • System and method for recognizing posture of single pig body based on stacked hourglass network
  • System and method for recognizing posture of single pig body based on stacked hourglass network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] As an aspect of the embodiment of the present invention, this embodiment provides a single pig body posture recognition system based on stacked hourglass network, refer to figure 1 , shows a schematic structural diagram of the single pig body posture recognition system based on the stacked hourglass network of the present invention, the system includes a data set processing module, a pig body posture estimation module based on the stacked hourglass network, and a pig body posture recognition module. Among them, the data set processing module 1 is used to establish an available pig body posture data set; the pig body posture estimation module 2 based on the stacked hourglass network is used for different selections according to the set model training parameters and model frameworks (based on different hourglass models). The connection mode between the structures and the hourglass structure participating in the connection) are used for training operations, and the trained ...

Embodiment 2

[0034] As another aspect of the embodiment of the present invention, this embodiment provides a single pig posture recognition method based on stacked hourglass network, refer to figure 2 , shows a flowchart of a single pig body posture recognition method using the stacked hourglass network-based single pig body posture recognition system of the present invention.

[0035] The first is the data preprocessing process. This step requires that the image data contains only one pig and is located in the middle of the image as far as possible, and that it is acquired through the side view angle as much as possible. The image data set is composed of image and video data obtained by manual use of SONY SELP1650 (25fps) equipment (resolution 1920×1080) and images obtained by web crawler technology. The video data needs to be divided into video frames Processing (set the unit interval to 1s) and structural similarity analysis, and manually screen the images obtained directly and throug...

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 provides a system and method for recognizing the posture of single pig body based on a stacked hourglass network. The method comprises the steps: acquiring image data and video data on the spot through photographing equipment, acquiring the image data through a web scraping technology, learning a preprocessed data set through the training of an improved stacked hourglass network model, automatically acquiring a skeleton diagram of a single pig in a pigsty within a shooting controllable range, and determining the category of the pig according to obtained pig body posture data; andwhen a certain posture of the pig is abnormal, warning and prompting a breeder to enter the pigsty to further check the health condition of the specific pig. With the system and the method, pigs in afarm can be effectively managed, and the pressure of technicians such as feeders is relieved to a certain extent.

Description

technical field [0001] The present invention relates to the field of image data processing and deep learning, in particular to a single pig posture recognition system and method based on a stacked hourglass network. Background technique [0002] China is a big pork consuming country as well as a big breeding country. Pig farming is one of the important pillar industries of animal husbandry and occupies an important position in agricultural production. According to the report of the National Bureau of Statistics, the output of pig, beef, mutton and poultry meat in 2018 was 85.17 million tons, and pork consumption accounted for about 63% of the total meat consumption, which is enough to illustrate the importance of pig farming in the livestock and poultry farming industry. [0003] In recent years, with the development of society and the continuous improvement of people's living standards, higher requirements have been put forward for the quality of life; people have also beco...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06Q50/02
CPCG06Q50/02G06V40/20G06N3/045G06F18/22G06F18/214G06F18/2415
Inventor 王敏娟张雯欣杨正洪王兵兵
Owner CHINA AGRI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products