Method and device for extracting eyelash image and locating pupil through cross neighborhood method

Abstract

The invention relates to a method for extracting an eyelash image and locating a pupil through the cross neighborhood method. The method for extracting the eyelash image and locating the pupil comprises the following steps that gray processing is conducted on an eye image so that a gray image can be obtained and pixel points, in a cross neighborhood of each pixel point, of the gray image are ranked from large to small so that a plurality of pixel sequences can be obtained; according to a first difference value, the eyelash image is obtained; according to a second difference value, a pupil boundary image is obtained; binarization processing is conducted on the pupil boundary image, so that a pupil binarization image is obtained; a horizontal projection curve and a vertical projection curve are obtained according to the pupil binarization image; a vertical filtering curve and a horizontal filtering curve are obtained according to the horizontal projection curve and the vertical projection curve; the position of one pupil is determined according to a first half-height point, a second half-height point, a third half-height point and a fourth half-height point. According to the method for extracting the eyelash image and locating one pupil through the cross neighborhood method, the calculated amount can be effectively reduced, the processing speed can be improved, and noise in the eye image can be efficiently eliminated.

Description

technical field [0001] The invention relates to a method for extracting eyelashes and locating pupils, in particular to a method and device for extracting eyelash images and locating pupils by a cross neighborhood method. Background technique [0002] Biometrics is a technology that uses biological characteristics to identify an individual. Among all biometrics, iris recognition is the most accurate, safest, most stable, healthiest and most convenient. Iris-based automatic identification and verification systems have grown in popularity over the past few years due to the advent of efficient, available and accurate iris recognition algorithms. Iris recognition includes six steps: image capture, image preprocessing, iris location, feature encoding and feature matching. Pupil positioning is a part of iris positioning. The quality of this step will directly affect the noise capacity of the iris recognition system, and eventually lead to the limitation of the application field ...

AI Technical Summary

Problems solved by technology

However, in the existing technical solutions, all methods with better positioning effects are two steps from rough positioning (image binarization processing) to fine positioning (least squares fitting or circular edge detector or operator edge processing) The way to go, this method is complex, the amount of calculation is large, and for noisy images, the effect of these methods is not satisfactory

Images