Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Behavior quantification method based on states and maps and terminal

A behavioral and graph technology, applied in the field of data analysis, can solve the problems of staying, quantitative analysis of unavailable results, accurate classification errors, etc.

Active Publication Date: 2020-12-11
深圳市一湾生命科技有限公司
View PDF12 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The high-dimensionality and complex structure of animal behavior also blur the boundaries between different behaviors, reduce the differences between different behavior categories, and bring a lot of errors to the accurate classification of classifiers such as SVM
Although this method can realize the recognition of various moods of animals, it does not give more quantitative analysis of the results, and still stays in the traditional counting and statistical method

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
  • Behavior quantification method based on states and maps and terminal
  • Behavior quantification method based on states and maps and terminal
  • Behavior quantification method based on states and maps and terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] The embodiment of the present invention discloses a behavior quantification method based on state and map, such as figure 1 shown, including the following steps:

[0028] Step 101, dividing the original behavioral data into segments to generate stateful behavioral data;

[0029] Specifically, the "segmenting the original behavior data into segments to generate state-based behavior data" in step 101 includes: tracking the preset target based on the original behavior data and the preset pose estimation method (that is, the original behavior data A plurality of body feature points corresponding to the target) to generate a time series related to the feature points; a dynamic time series segmentation algorithm is used to segment the time series to generate stateful behavior data. Further, the dynamic time series segmentation algorithm includes: a hierarchical alignment clustering analysis algorithm.

[0030] Specifically, through step 101, the behavior data is segmented, ...

Embodiment 2

[0043] On the basis of Embodiment 1, Embodiment 2 of the present invention also discloses a behavior quantification method based on state and map, specifically, as figure 2 As shown, multiple units can be set in this scheme, specifically including: animal behavior fragmentation unit 201, animal behavior fragment feature extraction unit 202, behavior low-dimensional mapping unit 203, behavior state clustering unit 204, behavior state map analysis unit 205 and behavioral state transition analysis unit 206, these six parts.

[0044] In this way, step 101 in Embodiment 1 is executed by the animal behavior fragmentation unit, step 102 in Embodiment 1 is executed by the animal behavior fragment feature extraction unit, step 103 in Embodiment 1 is executed by the behavior low-dimensional mapping unit, and step 103 in Embodiment 1 is executed by the behavior low-dimensional mapping unit. The behavior state clustering unit executes step 104 in Embodiment 1, and the behavior state grap...

Embodiment 3

[0052] Embodiment 3 of the present invention, on the basis of Embodiments 1 and 2, proposes a behavior quantification method based on state and map based on specific application scenarios. The specific process is as follows:

[0053] First, the animal behavior fragmentation unit 201 is used to divide the raw animal behavior data into fragments in the time dimension. Animal behavior data is generally collected by a camera, which is video data. The behavior of animals is related to the movement of key parts of the body. Therefore, multiple body feature points of the target animal can be tracked through animal pose estimation methods (for example, existing animal pose estimation methods can be used), and animal motion skeletons can be constructed to generate and feature points. Correlated high-dimensional time series. The high-dimensional time series has time dynamics, so the dynamic time series segmentation algorithm can be used to fragment the animal motion skeleton. The spec...

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 provides a behavior quantification method based on states and maps and a terminal. The method comprises the steps of segmenting original behavior data into segments, so as to generate stateful behavior data; extracting a dynamic nucleation distance from the behavior data to construct a feature matrix; performing low-dimensional mapping on the feature matrix to generate a behavior map; determining an optimal behavior state clustering number based on a preset clustering effect evaluation module, and clustering behavior states in the behavior map based on the optimal behavior stateclustering number; and carrying out quantification and visual display on the spatial information and the time information in the clustered behavior maps. According to the behavior quantification method based on the state and the maps, high dimensionality and diversity of behaviors are reserved in a low-dimensional space, so that finer indexes can be set and detected in the same experimental paradigm, more reference data are provided for research and development of neuropsychiatric drugs, and the efficiency and accuracy of drug effect detection are improved.

Description

technical field [0001] The invention relates to the technical field of data analysis, in particular to a behavior quantification method and terminal based on state and map. Background technique [0002] In the research and development of neuropsychiatric drugs, the behavioral differences of model animals before and after drug use are important indicators for judging drug efficacy. For example, in the research and development of anxiety (Anxiety) drugs, the elevated plus maze test (Elevated Plus Maze, EPM) or the open field test (Open Field Test, OFT) is often used to judge the anxiety level of the experimental mice before and after the drug. [0003] Among them, the EPM is composed of two open arms (Open Arm) and two closed arms (Close Arm), which are crossed in a cross shape, and the cross part is the central area. The entire cross-shaped maze has a certain height from the ground. Mice are curious and want to explore facing the open arm, and at the same time have the natur...

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
IPC IPC(8): A61B5/11A61B5/00
CPCA61B5/11A61B5/4848A61B5/7264
Inventor 韩亚宁黄康蔚鹏飞王立平
Owner 深圳市一湾生命科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com