Skip to Main Content
In this paper, we propose a real-time hand gesture recognition system using a thinning method utilizing a stereo camera. We implement a depth map in the hand detection portion that uses a sum of absolute differences method based on the acquired right-left image to detect the foreground object. We use a convex hull to detect the region of interests (ROI) and calculate the depth of object in ROI to obtain hand images that are more accurate. Then, we remove the background image in ROI and get the foreground image as a hand image. Finally, we use a blob labeling method to obtain the clean hand image, without the noise caused by computing distance from stereo matching. The hand gesture recognition system uses the Zhang and Suen thinning algorithm to obtain the feature point, angle and distance. It recognizes five kinds of hand gestures. The proposed method achieves an average recognition rate of 82.93%.