Spelled sentence recognition using radon transform

Various sign languages are used in India, but in schools for deaf, American Sign Language (ASL) is taught. So, the work is based on ASL. Sign recognition application is the development of more effective and friendly interfaces for human-machine interaction. It can provide an opportunity for a mute person to communicate with normal people without the need of an interpreter. We propose a novel system for recognition of spelled sentences from a video, based on radon transform. An algorithm is used to separate out key frames, which contain correct gestures from a video sequence. Segmentation is applied on key frames to separate out hand from complex and nonuniform background. Features are extracted by radon transform and gesture is recognized.