Computer Vision Based Human-Computer Interaction Using Color Detection Techniques

A gesture-based human computer interaction allows people to control the application on windows by moving their hands through the air and make computers and devices easier to use. Existing solutions have relied on gesture recognition algorithms they needs different hard wares, often involving complicated setups limited to the research lab. Algorithms which are used so far for gesture recognition are not practical or responsive enough for real-world use, might be due to the inadequate data on which the image processing is done. As existing methods are based on gesture recognition algorithms. It needs ´ANN training´ which makes whole process slow and reduces accuracy. Method we proposed is based on real time controlling the motion of mouse in windows according to the motion of hand and fingers by calculating the change in pixels values of RBG colors from a video, ´without using any ANN training´ to get exact sequence of motion of hands and fingers.