Insect dynamic behavior identification method based on deep learning and image technology

Abstract

The invention relates to a behavior recognition method, in particular to an insect dynamic behavior recognition method based on deep learning and an image technology. The identification method comprises the following steps: 1) acquiring a data source video; 2) processing the image; 3) establishing an identification model; 4) extracting time and space features; 5) performing deep learning; and 6) identifying and classifying the dynamic behaviors of the insects. According to the identification method, the insect body is divided into the head region and the tail region, the ROIs are extracted respectively, and ROIs of the head region and the tail region are detected respectively, so that misjudgment caused by vibration interference of the current behavior of the insect on other parts of the body can be effectively avoided, and the accuracy can be effectively improved; meanwhile, after the key point recognition model and the neural network model are trained, the insect behavior generation result can be automatically recognized through the key point recognition model and the neural network model, and then the manual observation time can be greatly shortened; the problems that an existing method is time-consuming, labor-consuming and poor in accuracy are solved.

Description

technical field [0001] The invention relates to a behavior recognition method, in particular to a dynamic behavior recognition method of insects based on deep learning and image technology. Background technique [0002] Crops and stored products have always been attacked by pests. Studying the types, laws and functions of insect behavior can provide a theoretical basis for insect control and forecasting. In recent years, computer vision technology has been applied more and more in agricultural production. Using computer instead of manual monitoring and statistics has higher efficiency and provides reliable and accurate basis for the regulation of agricultural production. [0003] At present, there are few methods specifically for insect behavior recognition. The behavior recognition of animals is mainly through manual observation, analysis and statistics, optical flow method and key point positioning method. When manually observing and analyzing statistics, researchers need...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a deep learning-based method that can greatly reduce the time of manual observation while ensuring the accuracy of insect behavior detection and analysis, so as to solve the problems of time-consuming, laborious and poor accuracy of existing methods. Insect Dynamic Behavior Recognition Method Based on Image Technology

Images