Hand-gesture recognition for automated speech generation

In the past few years the development in the field of Gesture Recognition and the tools, devices, models and algorithms are into existence. Regardless of various innovations in the above technologies humans are still comfortable with showing gestures according to the plain hand. In this paper we focus on a system that runs on a mobile computing device which provides us the technology for automated translation of the Indian Sign Language system into Speech in the English Language and thus provides a duplex communication between the vocally impaired people and the common people of the society. This system can be very effective and can be used in the near future for communication between the people who do not understand the Sign language since it works on the principle of Gesture Recognition. The system uses an intrinsic mobile camera for gesture recognition and acquisition; gesture acquired is processed with the help of Algorithms like HSV model-(Skin Color Detection), LargeBlob Detection, Flood Fill and Contour Extraction. The system is able to recognize one handed sign representation of the standard alphabets (A-Z) and numeric values (0-9). The output of this system is very efficient, consistent and of high approximation of gesture processing and speech analysis.