查看文章

We present in this paper a new approach for communication with computer systems using hand gesture. The process is based on extracting features of the hand region and then using these features for communication with computer system. Our approach is defined in three steps (i) Segmentation for hand localization,(ii) Feature extraction and (iii) Feature identification to create a fuzzy rule base. The gesture is a natural way of communication for many people with sign language. In computer systems, gesture recognition is very difficult and complex task since the full recognition system should be able to identify the hand in different scales, positions, orientations, contrasts, funds, luminosity, and others. The gesture communication with machines are enriched by instrumented gloves, which are used to control virtual actors, to describe and manipulate objects on computer screens or even to recognize the sign language. Unfortunately, these gloves are expensive and fragile, and their cables are a hindrance. We present a new system using hand detection technique based on the color classification using fuzzy logic system, image segmentation using edge detection and use Ellipse method and Comparative based method to localize the hands from others objects detected in the image.