Body-Part Tracking for Individual Mouse Identification in Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technologies fail to identify individual animals accurately when they are moving or in groups, requiring costly equipment like cameras and thermographs, and lack applications beyond health management.
Innovation Solution
An information processing device that uses image acquisition, part extraction, individual identification, and model selection to analyze animal activity, employing skeleton estimation and individual identification models to identify and track animals as individuals in a moving state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional technology is used to measure biometric data contactlessly, then biometric data can be measured without contact, but the animal cannot be identified as an individual when moving or in groups
Solution Approach 1:
The patent segments the animal body into multiple parts (head, body, limbs, tail) and tracks each part's movement independently across multiple images. This segmentation allows the system to distinguish individual animals based on their unique movement patterns even when they are moving or in groups, resolving the contradiction between contactless measurement and individual identification.
Solution Approach 2:
The patent performs preliminary actions by extracting skeletal information and movement trajectories before final individual identification. The system pre-processes images to obtain skeletal structures and movement patterns, which are then used to identify individuals, enabling both contactless measurement and individual differentiation.
2Reliability
If camera and thermograph are used to check health condition, then health monitoring is possible, but device cost increases
Solution Approach 1:
The patent makes the image processing device universal by enabling it to perform multiple functions: biometric data measurement, individual identification, and health condition monitoring. By using a single device for multiple purposes, the system eliminates the need for separate expensive thermographs while maintaining comprehensive monitoring capabilities.
Solution Approach 2:
The patent creates a virtual copy of the animal's skeletal structure and movement patterns from 2D images, allowing health monitoring without physical contact or additional thermal imaging equipment. This virtual representation enables health assessment through movement analysis alone.
3Measurement precision
If multiple skeleton estimation models and individual identification models are used, then identification accuracy is improved, but model selection complexity increases
Solution Approach 1:
The patent changes parameters such as image resolution, skeletal extraction thresholds, and movement detection sensitivity to optimize identification accuracy. By adjusting these parameters rather than using multiple complex models, the system achieves high accuracy while keeping the selection process simple.
Data Source
AI summary
A challenge is to provide an art to identify mice as individuals from captured images of one or more mice that are doing activity in a predetermined activity range. An image processing device includes an image acquisition unit, a part extraction unit, and an individual identification unit. The image acquisition unit acquires a captured video of a state in which one or more animals are doing activity in a predetermined activity range. The part extraction unit extracts multiple parts of the body of each of the one or more animals from each of unit images included in the video. The individual identification unit analyzes the parts extracted from the unit images by the part extraction unit in a time-series manner and identifies the one or more animals as individuals in each of the unit images on the basis of the analysis results. Thus, the challenge is solved.