Live pig behavior identification method and device based on video images

Abstract

The embodiment of the invention provides a live pig behavior identification method and device based on video images. Target tracking is carried out on a live pig and the behavioral characteristics ofthe live pig is extracted, the method has the characteristics of high speed and high detection rate; real-time processing of video tracking, HOG, HOF, MBH feature and the Dense Trajectory feature candescribe the bottom semantic feature of the behavior of the target to a great extent; the Naive Bayes nearest neighbor algorithm (NBNN) effectively avoids quantization errors caused by feature quantization in the visual word bag model, can naturally process the classification problem of a large amount of data, and effectively avoids overfitting.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of video image processing, in particular to a method and device for recognizing pig behavior based on video images. Background technique [0002] my country is the largest producer and consumer of pork in the world. In recent years, the scale of pig farming in my country has continued to expand. Although the pig breeding industry has developed rapidly, the quality and efficiency of the pig breeding has not been improved. Compared with developed countries, the human cost of pig breeding in my country is too high. The behavioral characteristics of pigs, such as feeding, drinking, lying still, excreting, walking and running, reflect the growth status of pigs. At present, the analysis of the health status of pigs in my country mainly depends on manual observation. This work is time-consuming and laborious. The cost is not conducive to the efficient production of pigs. [0003] Existing vid...

AI Technical Summary

Problems solved by technology

Although the pig farming industry has developed rapidly, the quality and efficiency of the pig farming has not been improved. Compared with developed countries, the human cost of pig farming in my country is too high

The behavioral characteristics of pigs, such as feeding, drinking, lying still, excreting, walking and running, reflect the growth status of pigs. At present, the analysis of the health status of pigs in my country mainly depends on manual observation. This work is time-consuming and laborious. Cost, not conducive to efficient pig production

[0003] Existing video image recognition technologies mainly focus on human behavior recognition, and there are few systematic methods mainly aimed at pig behavior recognition

Images