Gesture reconition using depth image

In this paper, we propose and implement a novel method for recognizing gestures using depth image from Microsoft´s Kinect sensor. Our approach consists of two parts: gesture segmentation and gesture classification. For gesture segmentation, we use a histogram based threshold method to segment gestures. We also used median filter for image filtering to remove noise. For gesture classification we used corrosion operation to find the palm center, and we devise a method based on Graham´s Scan algorithm to find fingertips. The experimental results indicate that our method can effectively recognize gestures in different lighting conditions and could be used in real-time video processing.