The Human Eyes-changing Recognition System Based On Computer Image Processing Technology

With the socio-economic development, Motor vehicles increased, traffic accident caused by fatigue driving is increasing .It is very important to study a set of effective, safe and reliable anti-fatigue warning system. Many countries are very active to research an anti-fatigue warning system. The eyes changed recognising system based on image processing is a part of anti-fatigue warning system.This paper have a deeply research on human detecting, human eyes recognising and locating, improve on the Hsu RL complexion detecting algorithm, detecting the complexion region in allusion to each image according its complexion distributing, then combined with brim detecting to make a accurate locating Resolve effective former algorithm that detect over.According to human eyes recognition system based on computer-based image processing technology, we need to track the eyes, locate the eyes and make a presented judge on the eyes state. Rough positioning by eyes blink detection, using the improving Susan corner extracting operator to determine the precise location within the corners of human eyes ,using the features of the inside corner of the eyes to dynamically track the eyes, in tracking process automatically generated eyes open template, using the template and current eyesregion to determine open, closed state ,According to the change of tired eyes, using the information of the blinking frequency Further determine whether the driver is fatigue.Using the SEED-DM642 with strong image processing as a development platform, the system is programming and debugging in CCS. The experiments demonstrate, the system can accurately extract the eyes character, detecting the eyes state with open and close。The system have fine practicability and reliability, it can effective detected if the driver is in fatigue state.