A Novel Approach for Pupil detection

摘要：

This paper presents a simple and novel approach for segmentation and detection of pupil which is fast and efficient,useful for IRIS recognition system related to IRIS based Biometric systems.The current pupil detection approaches to detect the pupil are computationally costly.Furthermore,the current approaches sometimes dont have sufficient accuracy and the mechanism to deal with spot reflection issues sometimes leading to errors.In this paper,we deal with the said problems and propose a simple and novel approach to detect the pupil accurately and eliminate/minimize the noise.At first,the proposed Power based Logarithmic enhancement Algorithm (PLEA) algorithm is applied,then Gaussian filtering is applied on the given processed image and later on,it is subjected to Sobel Edge detection and then morphological image cleaning process and in the last step,the pupil detected is marked,the level of efficiency of the detection marked is gauged using Pratts figure of merit

关键词： Mathematical morphology Sobel Edge detection segmentation Gaussian feature extraction Iris Pupil detection segmentation Morphological Image Cleaning

作者: Brhmadesam Sateesh

作者单位: Research & Development Centre,Samsung India Electronics Pvt Ltd,India

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 317-321

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）