Mouse characteristic behavior analysis method based on scene geometric constraint and deep learning

Abstract

An experimental mouse characteristic behavior analysis method based on scene geometric constraint and deep learning belongs to the technical field of machine vision, and comprises the following steps: configuring markers, extracting scene information, and constructing a structured scene space; collecting mouse video data, and creating a data set; constructing a mouse semantic posture model by using the semantic modeling points; a deep neural network is selected for training and learning, and dynamic information of mouse semantic statue points is extracted; based on a view geometry principle, mapping mouse semantic modeling points of a video image to a real scene to realize regression positioning of the modeling points; using a density clustering algorithm to analyze the regression positioned molding points, and locking key sequence frames; under different plane spaces, characteristics and differences of behaviors are analyzed according to field geometry constraint conditions. Finally, scene geometric constraint conditions are established through a structured space scene, complex and high-dimensional behaviors are changed into low-dimensional physical quantities, and interference-free fine-grained recognition and quantitative analysis on the behaviors of the experimental mice are achieved.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a mouse characteristic behavior analysis method based on scene geometric constraints and deep learning. Background technique [0002] The characteristic behavior of mice is used to reveal the potential strategies of the brain to adapt to the environment, to capture changes caused by genetic or neural manipulation, is one of the behavioral methods to assess learning and memory ability, is an indicator to measure medication, mood, etc., is a rodent It is an important part of animal behavior and pharmacology research, has an irreplaceable role, and has broad research prospects and guiding significance. A recent trend in laboratory mouse behavior research is to capture activities and behaviors in the natural environment. However, the typical behavior of laboratory mice is mainly studied by human eyes or by sticking markers and implanting sensors on the mouse. And using infrar...

AI Technical Summary

Problems solved by technology

A recent trend in laboratory mouse behavior research is to capture activities and behaviors in the natural environment. However, the general research on the characteristic behavior of laboratory mice is mainly through human eye observation or labeling mice, implanting sensors And using infrared devices to measure, these methods affect the normal activities of mice to a certain extent, keeping the movement of mice in an unnatural state

Images