Fish attack behavior identification and quantification method and device under group background and storage medium

Abstract

The invention discloses a fish attack behavior identification and quantification method and device under a group background and a storage medium, the fish attack behavior identification and quantification method under the group background comprises the following steps: a sampling step: collecting a fish motion sample; an information extraction step: extracting the swimming speed, direction and relative position coordinates of a single target individual; a motion information calculation step: calculating to obtain motion characteristics of the multi-target fish school; a network construction step: constructing a fish interactive social network topological structure model under the group background; a feature calculation step: calculating typical feature parameters of attack behaviors through the network topology structure model; and an attack intense degree judgment step: judging the state and degree of the fish school attack behavior according to the typical characteristic parameters. The method can provide a basis for monitoring abnormal movement of fish schools in actual production, reducing fish attacks in the culture process and improving culture benefits, and is beneficial to healthy development of the culture industry.

Description

technical field [0001] The invention relates to a method, a device and a storage medium for identifying and quantifying fish attack behavior in a group background. Background technique [0002] Fish, like other animals, have aggressive behaviors to protect resources such as their food or territory, which is also a distinctive feature of fish. In a high-density breeding environment, due to stress and pressure, many fish will behave aggressively. During the breeding process, a large number of individuals with few fins and tails can be seen, resulting in slow growth, decreased immunity, and even a large number of deaths, causing great harm to farmers. Economic losses. Monitoring fish aggression can provide early warning of fish welfare status without adverse effects on fish. [0003] How to identify and quantify fish attack behavior under aquaculture conditions is a difficult problem in production management and a key problem to be solved in order to realize welfare farming. ...

AI Technical Summary

Problems solved by technology

[0005] 1. There are few studies on the identification and quantification of fish aggressive behavior characteristics in the group background;

[0006] 2. The social network analysis in the process of fish attack in the group background is almost blank

Images