Static Hand Gesture Recognition Based On Hand Shaped Features

With the development of the computer techniques, gesture recognition is becoming one of the key techniques of HCI. It is a hot research topic in the fields of image processing, pattern recognition, computer vision etc. However, hand gestures recognition is an extremely challenging inter-disciplinary project, due to two reasons: firstly, hand gestures are rich in diversities, multi-meanings, and space-time varieties; secondly, human hands are complex non-rigid objects.This paper analyses the ways of hand gesture recognition and designs a system of real-time hand gesture recognition. Four modules are included in the system: hand gesture collecting, image pre-processing, feature extraction, and gesture recognition. The system runs in the following steps: first, collect the hand gesture images with USB-camera; second, get the binary images and contour images through image pre-processing; third, extract features of the two gesture images; and at last, recognize the hand gesture based on template matching of Euclidean distance. OpenCV library is added in the course of designing the two system modules: image preprocessing and feature extracting, so a good performance is gained in both the running efficiency and speed. When testing the system, a gesture template library is set up for 10 kinds of hand gestures. 687 samples are recognized under two different distance conditions and the average recognition rate is 92.80%. The result shows that the system is good enough to meet the needs of real-time reaction and high recognition rate.