摘要
Iris localization is an important part of iris recognition. Aiming at the influence of irregular pupil circle and illumination inequality in dynamic iris localization in actual scene, a dynamic iris inner edge location method based on distance regularized level set model was proposed. According to the geometric features of iris, the pupil and iris are coarsely located through the secondary projection of binary image, and the initial contour of evolution of horizontal set is determined. Accurate positioning of the inner edge of iris is achieved by the improved distance regularized level set model of edge indicator function. Iris images Database were respectively experimented with this paper and another improved distance regularized level set model, and the positioning effect was evaluated by using the Jakard similarity coefficient. The results show that the distance regularized level set model based dynamic iris internal edge localization algorithm proposed in this paper can significantly reduce the number of iterations and positioning time of the localization algorithm on the basis of ensuring the positioning accuracy,which provides a theoretical basis and technical reference for the application of iris recognition technology.
学科与关键词
- Challenges, taxonomy and techniques of iris localization: A surveyGautam, Gunjan*; Mukhopadhyay, SusantaDigital Signal Processing: A Review Journal , 2020, 107: 102852.被引 4
- 基于统计特征的非理想虹膜图像分割方法万洪林; 李宝生; 韩民光电子.激光, 2013, (12): 2383-2391.
- An OCaNet Model Based on Octave Convolution and Attention Mechanism for Iris RecognitionZou, Dong; Feng, Jianbing*; He, Zhixin; Liu, Liping; Zhao, Meijun; Zheng, LizhongMATHEMATICAL PROBLEMS IN ENGINEERING, 2021, 2021: 3412060.被引 1
