Cyber Times International Journal of Technology & Management

The recent work in technology has changed the means of interaction between the digital world and real time applications. The main working of human computer interaction is to develop better intercommunication between user and computer by making computer more receptive to the user’s needs. Gesture is a powerful form of communication among humans. This paper represents simple as well as effective method of realizing hand gesture recognition. Our aim is to implement a system for hand gesture recognition that enables interaction with computer without attaching any  hardware. The methodology used is based on the Haar Cascade Classifier where the user will be able to detect the gestures stored and PCA (Principal Component Analysis) for gesture recognition. The images are a subset of new set of Sign Language. Eigen values of the Eigen Vectors are calculated. A gesture recognition pattern system is used to transform an image into feature vector called the averaged Image which will then be compared with the trained set of gestures. The method used was successful to retrieve the correct match. Our focus is to use this hand gesture detection and recognition easily accessible to the dumb and deaf user.