A New Robust Approach for Real-Time Hand Detection and Gesture Recognition

Gesture represents a non-verbal means of communication between people. It comes from poses or movements performed by the human body. Nowadays, hand gestures are the new modality for human and computer interaction. In real-time applications based on hand gestures, the main challenge is to detect the hand(s). In this paper, we present and analyze existing approaches for real-time hand detection. Then, we propose a new approach based on skin color detection. We try to reduce as much as possible the constraints and limitations of the existing approaches. Our method for hand segmentation detects the user's hand(s) even if the user's face or other peoples are viewed by the camera, and can know if the user is doing a gesture using one hand or two hands. Also, our method is robust to the scene's colors and illumination conditions, as it corrects the video's colors and luminance before performing the hand segmentation. We also studied in this paper the performance of hand detection based on skin color detection using three colors spaces. We show that the RGB mask was the best one. Our system detects several gestures, and translate them into functional ones, then, allowing the user to manipulate a 2D image in a real-time manner.