ConvLSTM network-based Mouse open field experiment behavior analysis method

Abstract

The invention discloses a ConvLSTM network-based mouse open field experiment behavior analysis method. The method comprises the following steps: outputting a feature map containing mouse key point semantic information and position information by using a key point detection model; inputting a feature map sequence formed by adjacent frames into an identification classification model established based on a ConvLSTM network to realize behavior classification; and finally, correcting a misclassification result by using mode filtering so as to obtain behavior parameters in the mouse open field experiment. Automatic identification of animal behaviors and automatic calculation of animal behavior indexes are realized. The workload of researchers can be reduced, a quantitative behavior analysis method is provided for the researchers, and the objectivity of experiments is improved. Meanwhile, the refined behavior analysis method can help a researcher to capture some behavior modes which are difficult to perceive, and the reliability of an analysis result is improved. Meanwhile, compared with an existing mouse behavior recognition method, the method of the invention is higher in accuracy, is less affected by environment and illumination changes, and is higher in robustness.

Description

technical field [0001] The invention belongs to the technical field of ConvLSTM network and behavior recognition, and in particular relates to a method for analyzing the behavior of mice in an open field experiment based on the ConvLSTM network. Background technique [0002] Animal ethology is a discipline that studies the functions, mechanisms, development and evolution of various animal behaviors. It aims to reveal the genetic basis of animal behavior, the behavior mechanism in ecology, the ecological significance and evolutionary significance of behavior, etc. Animal behavior It can provide a basis for judging the psychological and physiological state of animals. In the process of animal behavior research, experimenters directly or indirectly affect animal behavior by changing the surrounding environment (light, sound, electricity, drug treatment, food induction, etc.) Central nervous system function and mental state, etc., reflect the overall state of the experimental a...

AI Technical Summary

Problems solved by technology

[0003] At present, the traditional measurement method of behavioral indicators is based on the statistical analysis of the results of manual observation, which is time-consuming and laborious. It not only cannot meet the requirements of some experimental projects that require long-term observation, but also cannot objectively evaluate and measure behavior. Subjective factors Greater impact

In the current animal behavior analysis methods, the focus is mainly on the movement parameters of experimental animals, that is, the measurement of indicators such as speed, distance, and movement trajectory, while there are relatively few automatic analysis methods for animal behavior. The method of animal behavior is still mainly based on manual active observation, which is labor-intensive and at the same time, there are great deviations in the analysis results

Images