论文信息 - Localizing the iris center by region growing search

Localizing the iris center by region growing search

A novel approach to localize human iris centers in facial images is proposed. From the estimated nose center of the detected face, the proposed method searches for the possible iris edge by gradually growing a rectangle, and then localizes the iris center in an expected region by searching for the pixel, which has the minimum average intensity. Experimental results show that this method has good performance for facial images under variations in pose and lighting direction. The speed of the method is notably fast. It has been successfully applied to a face recognition system for providing the eye locations for aligning faces.

Wen Gao | Shiguang Shan | Debin Zhao | Bo Cao

[1] Feipei Lai,et al. Region-based template deformation and masking for eye-feature extraction and description , 1997, Pattern Recognit..

[2] Hong Yan,et al. Locating and extracting the eye in human face images , 1996, Pattern Recognit..

[3] Alan L. Yuille,et al. Feature extraction from faces using deformable templates , 2004, International Journal of Computer Vision.

[4] Irfan A. Essa,et al. Detecting and tracking eyes by using their physiological properties, dynamics, and appearance , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[5] Harry Wechsler,et al. Detection of faces and facial landmarks using iconic filter banks , 1997, Pattern Recognit..

[6] Kongqiao Wang,et al. A hierarchical multiscale and multiangle system for human face detection in a complex background using gravity-center template , 1999, Pattern Recognit..