An identification method and device based on a dynamic iris image

Abstract

The embodiment of the invention provides a recognition method and device based on a dynamic iris image, and belongs to the technical field of iris recognition in the field of biological recognition. According to the identification method based on the dynamic iris image, a stable feature sequence generated by using a plurality of iris images to find a feature repetition area as a stable area is used as a retrieval code, and high exclusiveness is achieved, so that most iris templates in a database are eliminated, and the range is greatly narrowed. And during registration and identification, characteristics are extracted by using a plurality of iris images, so that iris characteristic loss caused by eyelashes, eyelids, pupil changes and the like can be made up, a relatively complete iris image and a relatively complete comprehensive template are obtained, and the comparison precision is improved. And by establishing the relationship between the pupil change trend and the feature change trend, the effectiveness of the feature can be better judged. And a repeated area is found through comparison of a plurality of irises, and the security level can be set more flexibly for the probability of recording the repeated area.

Description

technical field [0001] The invention relates to the technical field of iris recognition in the field of biometrics, in particular to a recognition method and device based on a dynamic iris image. Background technique [0002] The existing iris comparison method narrows down the comparison range by comparing them one by one, or establishes an iris index coding and retrieval mechanism, and performs iris matching within this range. However, the current iris index encoding is to extract some main and simplified features of the iris for encoding. Although the matching range can be narrowed, it still has high repeatability. Moreover, the normalized iris image contains some invalid parts, which adds difficulty to the process of iris index encoding and retrieval. Because the iris will be affected by factors such as eyelashes, eyelids, and pupils when the iris image is collected, iris features will be lost, resulting in a decrease in comparison accuracy. Therefore, how to solve the...

AI Technical Summary

Problems solved by technology

However, the current iris index encoding is to extract some main and simplified features of the iris for encoding. Although the matching range can be narrowed, it still has high repeatability.

Moreover, the normalized iris image contains some invalid parts, which makes the process of iris index encoding and retrieval more difficult.

Because the iris will be affected by factors such as eyelashes, eyelids, and pupils when the iris image is collected, iris features will be lost, resulting in a decrease in comparison accuracy.

Images