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.

CN122397644APending Publication Date: 2026-07-17

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Filing Date
2026-06-17
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing pet activity status recognition methods cannot achieve multi-source information fusion, resulting in poor recognition accuracy and susceptibility to interference from various factors.

Method used

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.

Benefits of technology

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.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

本发明公开了一种基于多源传感器融合的宠物活动状态识别方法,具体涉及宠物饲养技术领域,包括以下步骤:步骤一:为宠物配置多源传感器采集系统,所述系统包括运动类传感器、生理类传感器、环境类传感器以及视觉和声音类传感器中的一种或多种,运动类传感器采集宠物肢体运动姿态、运动幅度和方向变化,生理类传感器反映宠物生理状态变化。本发明通过采集宠物佩戴终端、环境终端及视觉和声音终端的多源数据,并对其进行时间同步、预处理、特征提取与融合建模,进而实现对宠物活动状态的精准识别,本发明能够有效克服单模态信息不足、环境干扰大及识别稳定性差等问题,具有识别准确率高、鲁棒性强及可扩展性好的优点。
Need to check novelty before this filing date? Find Prior Art