一种基于多源传感器融合的宠物活动状态识别方法
By using a multi-source sensor fusion method, accurate identification of pet activity status is achieved, solving the problems of low identification accuracy and susceptibility to interference in existing technologies. This improves identification accuracy and robustness, supports health management and early warning, and is adaptable to different pet breeds and environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-06-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing pet activity status recognition methods cannot achieve multi-source information fusion, resulting in poor recognition accuracy and susceptibility to interference from various factors.
A multi-source sensor fusion method is adopted, which collects data through motion, physiological, environmental and visual/audio sensors, and performs time synchronization, preprocessing, feature extraction and fusion modeling. A multimodal fusion network with attention mechanism is used to fuse the data, train the pet activity state recognition model, and output the pet activity state and its confidence level in real time.
It significantly improves the accuracy of pet activity status recognition, enhances robustness and adaptability, can effectively identify pet activity status in complex environments, supports health management and early warning, and has real-time and scalability.