研究目的
Investigating the feasibility of recognizing human hand gestures using micro-Doppler signatures measured by Doppler radar with a deep convolutional neural network (DCNN).
研究成果
The proposed method achieved a classification accuracy of 85.6% for ten gestures and 93.1% for seven gestures. The study demonstrates the feasibility of using micro-Doppler signatures with DCNN for hand gesture recognition, though further research is needed to address variations in signatures due to environmental and user differences.
研究不足
The study was conducted with a single participant in a controlled environment. Variations in micro-Doppler signatures due to different aspect angles and distances to the radar were observed, indicating potential challenges in uncontrolled environments.
1：Experimental Design and Method Selection:
Employed Doppler radar to obtain micro-Doppler signatures of ten hand gestures. A DCNN was used to classify the spectrograms.
2：Sample Selection and Data Sources:
Measured ten different hand gestures from a single participant using Doppler radar.
3：List of Experimental Equipment and Materials:
Bumblebee Doppler radar (Samraksh Co. Ltd.), which operates at 5.8 GHz.
4：8 GHz.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: Hand gestures were measured, and their spectrograms were analyzed. A DCNN was trained with 90% of the data and validated with the remaining 10%.
5：0%.
Data Analysis Methods:
5. Data Analysis Methods: The classification accuracy was evaluated using five-fold validation.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
