In order to enrich the diversity of interactive recognition methods, an interactive method of gesture recognition based on static vision is proposed. The static gesture images are captured by color camera in real time. The gesture is extracted based on FHOG features. The extracted eigenvalues are used as input of SVM multi-class classifier to recognize gesture actions. The gesture features are used to locate feature points to achieve the segmentation of gesture recognition and gesture recognition. The experimental results show that the system can recognize six common static gestures. The system has good robustness, with an average recognition rate of 95.31%, a rejection recognition rate of 9.37%, and an overall recognition efficiency of 90.63%.

Date Published: 31 January 2020
PDF: 5 pages
Proc. SPIE 11427, Second Target Recognition and Artificial Intelligence Summit Forum, 114273C (31 January 2020); doi: 10.1117/12.2553006

Published in SPIE Proceedings Vol. 11427:
Second Target Recognition and Artificial Intelligence Summit Forum
Tianran Wang; Tianyou Chai; Huitao Fan; Qifeng Yu, Editor(s)