Date of Publication :21st February 2018

Abstract: This paper discusses the segmentation of hand gestures using skin detection method based on the YCbCr color space. Here a face detection algorithm based on viola jones has been used to detect and delete the face. Thus the only area to detect based on skin color is that of the hand. After skin color detection and thresholding using rectangular selection over the input image the binary image received is further processed using image processing techniques like median filtering and morphological operation. Then robust tracking of the segmented hand contour using image moment concept further helps in localizing position of hand contour in the two dimensional space. These segmentation of hand gestures can be used in creating datasets of binary hand gestures and in recognition of hand gestures based on the machine learning, neural network, geometrical analysis of hand and so on in future.

Reference :