Method for automatically acquiring boundary coordinates of inner circle and outer circle of iris

Abstract

The invention discloses a method for automatically acquiring boundary coordinates of an inner circle and an outer circle of an iris, and the method comprises the steps: S1, inputting an original human eye iris image into a convolutional neural network, and obtaining a corresponding feature map; s2, carrying out up-sampling on the feature map output by each convolution layer to obtain a feature map with the same size as the original human eye iris image, and then carrying out splicing to obtain a spliced feature map; s3, respectively inputting the spliced feature maps into a first multi-layer perceptron and a second multi-layer perceptron, and predicting parameters required by parameterized active contour models of the iris outer circle and the iris inner circle; and S4, respectively inputting the predicted parameters into the corresponding parameterized active contour models to obtain boundary coordinates of the iris outer circle and the iris inner circle. According to the method, the boundary coordinates of the inner circle and the outer circle of the iris can be automatically obtained from the human eye iris image, the boundary of the inner circle and the outer circle of the iris can be accurately positioned, and accurate input parameters are provided for subsequent iris normalization.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a method for automatically obtaining the boundary coordinates of the inner and outer circles of the iris. Background technique [0002] At present, as a biometric identification technology, iris recognition is more secure than face and fingerprint identification, and it is also recognized as the most accurate and effective biometric identification method. [0003] However, in practical applications, many technical challenges are still encountered, especially in the iris images of people collected in some complex and uncontrollable scenes, because of many changes in illumination, distance, occlusion, etc. There is some difficulty in segmenting out the iris region. [0004] In the existing methods of iris segmentation based on deep learning, most of the methods realize the segmentation of iris region through the classification of image pixels. The problems of these metho...

AI Technical Summary

Problems solved by technology

[0003] However, in practical applications, many technical challenges are still encountered, especially in the iris images of people collected in some complex and uncontrollable scenes, because of many changes in illumination, distance, occlusion, etc. Difficulty segmenting out the iris region

[0004] In the existing methods of iris segmentation based on deep learning, most of the methods realize the segmentation of iris region through the classification of image pixels. The problems of these methods are: there are often problems of over-segmentation and under-segmentation, which cannot be clearly defined. Locate the inner and outer circle boundaries of the iris

[0005] However, there is currently no technology that can accurately locate the inner and outer circle boundaries of the iris, thereby ensuring the overall accuracy of iris recognition

Images