1. 智能家居移动机器人的人脸识别方法.
- Author
-
李 珣, 李林鹏, 南恺恺, 王晓华, and 张 蕾
- Subjects
- *
HUMAN facial recognition software , *HOME automation , *MOBILE homes , *PROBLEM solving , *COMPUTER vision , *ROBOTS , *MOBILE robots - Abstract
To solve the problem that individual RGB images are susceptible to background interferencein face recognition systems, a real-time face recognition and tracking solution combining depth images is proposed. Propose a tracking mode combining Hungarian algorithm and HMM algorithm to filter potential interference information. Finally, the robot platform is built. The single-target and multi-target recognition test experiments were carried out in the real scene. The experimental results show that the proposed face recognition scheme has an accuracy of 94.3% for real-time face recognition, 76 ms for head positioning, and an average recognition time of only 761.17 ms.The experimental results show that the method is robust while maintaining good accuracy, and can meet the needs of smart home mobile robot face recognition. [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF