Extraction and behavior analysis method of multi-pig body movement trajectory in group environment

Abstract

The invention discloses a method for extracting and analyzing the motion trajectory of multiple pigs in a group environment, comprising the following steps: improving a known single-target tracking algorithm, and proposing an algorithm for simultaneously tracking multiple pigs in a group environment; The coordinates of the center point of each frame of each pig draw the movement trajectory and instantaneous speed of each pig; according to the target frame obtained by the tracking of the pig body, extract the region of interest, segment the target outline of each pig, and judge the pig body The current morphological feature is standing or lying down; combined with the movement track, movement speed, and contour shape of the pig body, combined with the movement law of the pig body and expert advice, the movement behavior of each pig in each time period is judged, mainly including: pig body standing, Lying prone, walking slowly, running fast, abnormal restlessness. The invention shows how to monitor the behavior of pigs in a group environment by using an accurate and robust tracking algorithm, which can detect abnormal movements of pigs in real time and assist in judging the health status of pigs.

Description

technical field [0001] The invention relates to the field of computer vision, and more specifically, relates to a method for tracking multiple pigs and determining their behavior in a group environment. Background technique [0002] Monitoring the behavior of pigs is the key link to judge whether the pigs are in an abnormal state, and can assist in judging the health status of the pigs. At present, pigs are generally raised in a centralized breeding method. The breeding scale is large and the number is large, and the manual detection method is difficult to timely An abnormal state of the pig body was found. The present invention uses target tracking and video segmentation to monitor the multi-target pig body in a group environment in real time, and uses target tracking to obtain the movement track shape and motion state changes of the pig body; uses video segmentation to obtain the appearance profile of the pig body, and evaluates the posture of the pig body . Finally, the...

AI Technical Summary

Problems solved by technology

[0002] Monitoring the behavior of pigs is the key link to judge whether the pigs are in an abnormal state, and can assist in judging the health status of the pigs. At present, pigs are generally raised in a centralized breeding method. The breeding scale is large and the number is large, and the manual detection method is difficult to timely Abnormal state of the pig body was found

Images