Hand Gesture Recognition using Virtual Canvas

Computer vision based hand tracing can be used to interact with machines in a new inventive way. The input components of a normal computer system uses keyboard, mouse and joysticks. Which can be avoided using gesture recognition. Gestures are revelatory body movements involving physical moves of the �ngers, hands, arms, head, face, or body with the purpose of: 1) assigning meaningful information or 2) interacting with the environment. They make up one interesting small subspace of possible human motion which tells someone to move or act in a certain way. A gesture may also be seen by the environment as a compact technique for the information to be transfuse elsewhere and subsequently adapted by the receiver. In this paper, we designed a robust real-time hand gesture recognition method. In our method, first step is capturing an image using webcam and then apply pre-processing steps for extracting features. The second stage involves the classification of the features set to the appropriate gesture using the trained classifier that was computed in the setup stage.