A fast iris detection method based on convolutional neural network

Abstract

The invention discloses a fast iris detection method based on a convolutional neural network. The method comprises the following steps: (1) designing and training an iris detection network regressor model; (2) acquiring an iris image to be detected and performing image preprocessing (3) Input the image obtained in step (2) into the iris detection network regressor model of step (1) to obtain the probability of belonging to the iris, the offset of the center point and radius of the iris image; (4) In order to obtain a more accurate iris position , output the center point and radius of the iris whose probability of belonging to the iris is greater than the preset threshold in the result obtained in step (3), and then screen the center point and radius of the iris of the final image by the non-maximum suppression method. The obtained iris area corresponds to the circular extent. The technical solution adopts a compact convolutional neural network to regress the center point and radius offset of the iris image, obtains the iris area according to the circular feature of the iris, has fast detection speed and high accuracy, and plays an important role in improving the iris recognition accuracy.

Description

technical field [0001] The invention belongs to the technical field of image processing and biometric information identification, and in particular relates to a fast iris detection method based on a convolutional neural network developed by embedding a neural network algorithm into a simulator for human iris image detection. Background technique [0002] The development of modern society has put forward higher requirements for the accuracy, security and usability of human identification. Identity recognition is a common problem encountered in people's daily life. Things such as the need to prove one's identity and identify others' identities often occur in life. The traditional identification method relying on photos has fallen far behind the requirements of the times, and human beings must seek new ways of identification that are safer, more reliable, and more convenient to use. The identification technology based on biometric identification has the following advantages: i...

AI Technical Summary

Problems solved by technology

[0004] Traditional iris detection methods mainly use the texture and grayscale features of the surrounding area of ​​the iris, and use methods such as edge detection, threshold segmentation, or local feature pattern (LBP) to detect the iris area. The detection of such methods is vulnerable to light, noise, and image blur. Influenced by factors, the stability is not high

Images