A fast and accurate algorithm for detecting and tracking moving hand gestures

Abstract

Human vision plays a very important role in the perception of the environment, communication and interaction between individuals. Machine vision is increasingly being embedded in electronic devices, as cameras are used with the function of perceiving the environment and identifying the elements inserted in a scene. Real-time image processing and pattern recognition are processing intensive tasks, even with the technology of today. This chapter proposes a vision system that recognizes hand gestures combining motion detection techniques, detection of skin tones, and classification using a model based on the Haar Cascade and CamShift algorithms. The new algorithm presented is 29% faster than its competitors.