Fast hand detection and gesture recognition

A fast hand detection and gesture recognition method is proposed in this paper. To reduce the computation time, we employ symmetric mask-based discrete wavelet transform (SMDWT) to reduce the image resolution and then utilize the extracted characteristics to perform hand detection and gesture recognition. The proposed method reduces about 66% of the overall computation time. Experimental results show that the proposed real time hand detection and gesture recognition methods can detect fast and accurately, and can be implemented on embedded system easily. The average gesture recognition rate is around 97.5%.