一种基于RFID技术的非接触式动态手势识别方法及系统
By combining KL divergence and frame difference methods for gesture segmentation and utilizing a hierarchical random forest model to improve real-time recognition, the problems of low gesture segmentation accuracy and poor real-time performance in existing technologies are solved, achieving high-precision and efficient gesture recognition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-09-12
- Publication Date
- 2026-07-17
AI Technical Summary
Existing RFID-based contactless gesture recognition technologies suffer from low gesture segmentation accuracy and poor real-time performance. In particular, gesture segmentation relies on the sliding window size, which is difficult to optimize, and the computational efficiency of DTW and KNN algorithms is low.
A sliding window algorithm based on KL divergence combined with frame difference method is used for high-precision gesture segmentation. A hierarchical random forest model is used to improve the real-time performance of recognition. The multipath effect of RFID signals is used to establish a gesture recognition model to reduce computational complexity.
It achieves high-precision gesture segmentation and improves the real-time performance of gesture recognition, reduces user fatigue when wearing the device, avoids the risk of privacy leaks, and enhances the flexibility and accuracy of the interaction process.
Smart Images
Figure CN117473420B_ABST