Author : Mr. Sunil T. Rajguru, Mr. Kishor N. Shedge and Mr. Devidas S. Thosar

I have proposed in this paper a unique technique which is able to adapt to noisy images for eye gaze detection as the processing of noisy sclera images captured at a distance and on the move has not been extensively investigated. The blood vessels of sclera have previously been investigated as an useful biometric characteristic.Capturing this section of the eye with a normal came ra with pictures of visible distance instead of near infrared images has given rise to attention in research interest This technique involves a sclera template rotational alignment and a distance scaling method to minimize the error rate when audible eye images are taken prisoner at a distance and on the move. The proposed system is being tested and results are generated by extensive simulation in java.

Keywords: Biometrics, feature extraction, Gabor filter’s, kernel, linear discriminates analysis, pattern recognition, sclera recognition