Pupil location algorithm combing VFC Snake model with Grayscale Features
- 10.2991/meita-15.2015.127How to use a DOI?
- VFC Snake model, least square fitting algorithm, pupil location, Grayscale Features.
Pupil location is one of the key steps for iris location as pupil boundary corresponds to iris inner boundary. But uneven illumination affects the segmentation result heavily. The pupil region is the darkest part in most of the images, but it’s not the only part. The proposed algorithm is based on the grayscale features and the prior that the pupil of humans is approximate circle. Firstly, the eye image binarization is achieved and the coarse positioning of the pupil region based on the grayscale features is obtained. Then, the outer boundary of the coarse positioning region is taken as the initial contour of the VFC snake model and iterates for a few times. Thirdly, a round curve is gained by using least squares circle fitting algorithm for the curve contour points, and serve as the new initial contour to iterative again. Finally, repeating the third step until the curve moved to the pupil edge. A large number of experiments demonstrate that proposed algorithm gives accurate result for most of the CASIA-Iris-Lamp images.
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Peipei Lv AU - Zhiming Wang PY - 2015/08 DA - 2015/08 TI - Pupil location algorithm combing VFC Snake model with Grayscale Features BT - Proceedings of the 2015 International Conference on Materials Engineering and Information Technology Applications PB - Atlantis Press SP - 698 EP - 703 SN - 2352-5401 UR - https://doi.org/10.2991/meita-15.2015.127 DO - 10.2991/meita-15.2015.127 ID - Lv2015/08 ER -