A method and collar for pet posture recognition and health monitoring
By combining a lightweight classification model with temporal fusion processing of acceleration and angular velocity features, accurate identification and health assessment of cat postures are achieved in low-power devices. This solves the problems of misjudgment and insufficient closed-loop feedback in existing posture recognition algorithms, and provides accurate health monitoring information.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN SMART TRAVELER TECH CO LTD
- Filing Date
- 2026-04-30
- Publication Date
- 2026-06-30
AI Technical Summary
Existing posture recognition technology lacks optimization for cat-specific behavioral characteristics, making it difficult to operate efficiently in low-power embedded devices and failing to achieve closed-loop feedback for health management.
A lightweight classification model is adopted, combined with temporal fusion processing of acceleration and angular velocity features. Attitude recognition is performed using triaxial acceleration data and triaxial angular velocity data, and periodic health status assessment is conducted to achieve a complete closed loop from attitude acquisition to health assessment.
It improves the accuracy of posture recognition and the timeliness of health assessment, reduces the system's dependence on external computing power, adapts to the battery life requirements of low-power devices, and provides accurate health reference information.
Smart Images

Figure CN122296260A_ABST