Drowsiness Detection for Drivers Using Computer Vision
Drows iness detecti on s ystem is regarde d as an effective tool to reduce the number of road acc ide nts. This project proposes a non -intrus ive approac h for detecting drows iness in drivers , us ing Computer V is ion. The algorithm is c ode d on Ope nCV platform in Linux environment. Th e parameters cons idered to detect drows iness are face and eye detection, blinking, eye c losure and gaze. Input is capture d a nd live fed from a camera that supports night vis ion as well. The algorithm is Haar tra ined to detect the face and the eye from the incoming frame. Once the eye is detected, further coding is done to track the eye and a utomatica lly set a dyna mic threshold va lue. Depending on the va lues obta ined from each of the inc oming fra mes and deviations from the threshold va lues , eye lid c los ure/bl ink/ga ze is detecte d. Warning s yste m is des igned to a lert the driver. This system renders an effic ient solution to road accidents and the cost of de ve loping it into a real time s ystem is als o feasible whe n compared to the cost involve d in the manufacture o f car.
OpenCV , Naive Bayes Classifiers,Drowsiness detection.