Finger vein segmentation method and device based on R2U-Net and storage medium

Abstract

The invention discloses a finger vein segmentation method and device based on R2U-Net and a storage medium. In a coding and decoding unit, a regular forward convolution layer is replaced by a cyclic convolution layer with a residual unit, the residual unit of the cyclic convolution layer is beneficial to establishing a more effective and deeper model, the cyclic convolution layer unit comprises aneffective feature accumulation method, and better and stronger feature representation is guaranteed by feature accumulation of different time steps. Thus, it facilitates the extraction of very low levels of features that are essential to the segmentation of finger veins. In the neural network training process, on one hand, the center of each image is randomly selected to obtain sub-blocks for data expansion, and on the other hand, the grains extracted by combining six traditional methods with different weights are used as gold standards, so that the method can give full play to the advantagesof each traditional extracted grain and make up for the disadvantages.

Description

technical field [0001] The present invention relates to the technical field of neural networks, in particular to a finger vein segmentation method, device and storage medium based on R2U-Net. Background technique [0002] In recent years, as people have higher and higher requirements for the security and accuracy of biometric systems, biometric technology has received more and more attention. As one of many biometric identification technologies, finger vein recognition has become a current research hotspot due to its advantages of non-contact acquisition, live detection, difficult forgery, and low cost. The segmentation of blood vessels in finger vein images is a key step in vein recognition technology, and the quality of the segmentation effect directly affects the precision and accuracy of subsequent recognition. [0003] In practical applications, the captured images not only contain vein patterns, but also irregular noise, shadows caused by different thicknesses of fing...

AI Technical Summary

Problems solved by technology

In addition, finger veins change with temperature or physical conditions, making it difficult to accurately extract vein vessel details

Traditional image segmentation algorithms, such as Otsu algorithm, entropy algorithm, K-means algorithm and fuzzy C-means algorithm, cannot achieve good segmentation results due to the need for more thresholds for low-quality finger vein images

[0004] Although semantic segmentation methods based on deep learning have been successfully applied to tasks such as image classification, segmentation and detection, and have...

Images