Multi-cues eye detection on gray intensity image

This paper presents a novel eye detection method for gray intensity image. The precise eye position can be located if the eye windows are accurately d etected. The proposed method uses multi-cues for detecting eye windows from a face image. Three cues from the face image are used. Each cue indicates the positions of the potential eye window. The first cue is the face intens ity because the intensity of eye regions is relatively low. The second cue is based on the estimated direction of the line joining the centers of the eyes. The third cue is from the response of convolving the proposed eye var iance filter with the face image. Based on the three cues, a cross-validati on process is performed. This process generates a list of possible eye wind ow pairs. For each possible case, variance projection function is used for eve detection and verification. A face database from MIT Al laboratory, whi ch contains 930 face images with different orientations and hairstyles capt ured from different people. is used to evaluate the proposed system. The de tection accuracy is 92.5%. (C) 2001 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.