A method for analyzing the movement behavior of mice in open field experiments based on key point detection

A behavior analysis and key point technology, applied in image analysis, neural learning methods, image enhancement, etc., can solve problems such as difficult development and poor generalization ability, and achieve strong anti-interference ability, strong robustness, and fine behavior. The effect of analytical tools

Active Publication Date: 2022-06-07
郑州布恩科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the purpose of the present invention is to provide a method for analyzing the movement behavior of mice in open field experiments based on key point detection for the problems of large development difficulty and poor generalization ability in the traditional behavior analysis methods in the prior art. Inputting the collected video can realize the calculation of a series of behavioral indicators. The technology based on deep learning has strong robustness, and has strong anti-interference ability to the environment and light changes, which greatly simplifies the analysis of behavioral science. The flow of indicators, the behavioral analysis method is convenient, comprehensive and precise, and has a good application prospect

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method for analyzing the movement behavior of mice in open field experiments based on key point detection
  • A method for analyzing the movement behavior of mice in open field experiments based on key point detection
  • A method for analyzing the movement behavior of mice in open field experiments based on key point detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0073] A method for analyzing the motor behavior of mice in the open field experiment based on key point detection, such as figure 1 shown, including:

[0074] (a) Collect experimental video and determine the experimental area;

[0075] (b) The mouse key point detection model based on convolutional neural network detects the mouse's nose tip, left ear, right ear and tail root in the video image, and returns the key point coordinates;

[0076] (c) Calculate the movement distance of key points between adjacent frames, and filter out abnormal key points with a distance threshold;

[0077] (d) Clean the coordinates of key points, use Kalman filtering to predict the trajectory, and eliminate the influence of measurement errors;

[0078] (e) Determine the movement parameters in the mouse open field experiment from the calculated coordinates of the key points, and generate relevant graphic reports.

[0079] In the step (a), as figure 2 As shown in the figure, the camera 3 is use...

Embodiment 2

[0091] The difference between this embodiment and Embodiment 1: in step (a), the experimental video is collected, and the method for determining the experimental area is:

[0092] (a1) The camera 3 used is a USB zoom camera. The size of the captured image is 640*480 and the frame rate is 30. The camera 3 is installed on the upper part of the bracket 1 and is about 70cm away from the bottom of the open field experiment box 2. During the experiment, adjust the focal length of the camera 3 by adjusting the focal length of the camera 3. Align the square box in the image to the bottom of the open field experiment box 2 to determine the experimental area.

[0093] (a2) The size of the open field experiment box 2 used is 45*45*40cm, and the background color is white and light gray with high contrast with the mouse body color. During the experiment, the experimental equipment is placed in a dark and quiet environment, and warm colors are used to make up Lights 4 provide illumination. ...

Embodiment 3

[0096] The difference between this embodiment and Embodiment 1: refer to Figure 4-5 , in step (b), key point detection is performed on the video collected in step (a), and the key point detection model based on convolutional neural network has the following characteristics:

[0097] (1) The input of the mouse keypoint detection model is 640*480, the backbone network is a residual network with 5 residual blocks, the residual network is used for feature extraction and data dimensionality reduction, and the image is downsampled after the residual network 16 times output 40*30*2048 feature map, then use transposed convolution to increase the resolution, perform 4 times upsampling, and output 80*60*12 feature map, which are used for category prediction and coordinate offset respectively.

[0098] (2) The mouse keypoint detection model uses atrous convolution to expand the receptive field of the network and fuse multi-scale information. In the block4 module of the residual module,...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method for analyzing the movement behavior of mice in an open field experiment based on key point detection, comprising: (a) collecting experimental videos and determining the experimental area; (b) a mouse key point detection model based on a convolutional neural network, Detect the nose tip, left ear, right ear and tail root of the mouse in the video image, and return the coordinates of the key points; (c) calculate the movement distance of the key points between adjacent frames, and filter out the abnormal key points with the distance threshold; (d) check the key points The coordinates are cleaned, and the Kalman filter is used to predict the trajectory to eliminate the influence of measurement errors; (e) determine the movement parameters in the mouse open field experiment from the calculated key point coordinates, and generate relevant graphical reports. The method for analyzing the movement behavior of mice in open-field experiments based on key point detection in the present invention has strong robustness, and has strong anti-interference ability against changes in the environment and light, and greatly simplifies the process of analyzing behavioral indicators. Behavioral analysis means are convenient, comprehensive and precise.

Description

technical field [0001] The invention relates to the technical field of behavior detection, in particular to a method for analyzing the movement behavior of mice in an open field experiment based on key point detection. Background technique [0002] Animal behavior can be regarded as the body language of animals expressing psychology and physiology, the complex result of the interaction between genes and the environment, and the embodiment of the comprehensive functions of animals themselves. Animal behavior can provide a basis for judging the psychological and physiological states of animals. Since the 1970s, related research on animal behavior has attracted the attention of scientists from all over the world, and it has become an extremely active and important branch in the field of life science research. It is to observe the behavior and changes of animals under specific laboratory conditions, and then study the neurological functions, psychological processes and drug effe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/246G06T7/277G06V10/77G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06T7/248G06T7/277G06N3/084G06T2207/10016G06T2207/20081G06T2207/20084G06T2207/30241G06N3/045G06F18/213G06F18/214
Inventor 朱俊才王治忠徐正阳王松伟牛晓可杨庭瑞张彦昆
Owner 郑州布恩科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap