Low-exposure vein image enhancement method based on cross-scale feature fusion

A vein image and feature fusion technology, applied in the field of computer vision, can solve the problems of vein detail information loss, unclear vein veins, low image contrast, etc., and achieve the effect of enhancing the ability of representation learning

Inactive Publication Date: 2021-08-17
CHINA UNIV OF MINING & TECH
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a low-exposure vein image enhancement method based on cross-scale feature fusion, ...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Low-exposure vein image enhancement method based on cross-scale feature fusion
  • Low-exposure vein image enhancement method based on cross-scale feature fusion
  • Low-exposure vein image enhancement method based on cross-scale feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Step 1. Collect 2000 images of veins on the back of the human hand under normal exposure, and convert the images of veins on the back of the human hand under normal exposure into low-exposure images of the back of the hand one by one through the low-brightness image synthesis method. 4000 vein images with a pixel size of 256×256, 4000 vein images are used as a training set, such as figure 2 shown;

[0047] Step 2. Use the cross-scale feature fusion module to replace the convolutional layer in the basic residual module, and add an attention mechanism to obtain the vein cross-scale fusion residual block; according to the required image enhancement effect (image evaluation index: structural consistency SSIM and peak signal-to-noise ratio (PSNR), construct several vein cross-scale fusion residual blocks;

[0048] The three scale branches of the image are obtained through the local maximum pooling method to obtain low-level input features. Each branch uses the same convolu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a low-exposure vein image enhancement method based on cross-scale feature fusion. The method comprises the following steps: making a low-exposure vein image data set, and constructing a cross-scale feature fusion module; based on a channel attention mechanism, constructing a vein cross-scale fusion residual block by adopting a residual structure; connecting and stacking a plurality of residual blocks end to end to form a vein image cross-scale fusion model, constructing a test set, and inputting the low-exposure vein image in the test set into the vein image cross-scale fusion model to obtain an enhanced vein image. According to the multi-scale branch cross-scale feature information fusion method provided by the invention, spatial structure information between different scales of the vein image is fully utilized, the representation learning ability of the network model for detail information such as vein veins is enhanced, and the enhancement effect of the low-exposure vein image is improved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a low-exposure vein image enhancement method based on cross-scale feature fusion. Background technique [0002] Compared with other biometric features (such as fingerprint, iris, gesture and face), vein blood has the characteristics of anti-counterfeiting and easy acceptance, and has become one of the most popular personal identification methods. However, due to the limitations of vein acquisition equipment, the collected images often have problems such as dark background, low image contrast, unclear vein veins and loss of vein information details. [0003] Vein recognition requires more vein detail information, and traditional low-exposure image enhancement algorithms cannot solve the problem of vein information loss. Lv et al. proposed a multi-branch low-light enhancement network. The network loss function involves the structure, context and regional information of the image, wh...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T5/50G06K9/62G06N3/04G06N3/08
CPCG06T5/50G06N3/08G06T2207/30101G06T2207/20221G06T2207/20024G06T2207/20081G06T2207/20084G06N3/045G06F18/253
Inventor 王军韩淑雨潘在宇李玉莲申政文陈晓玲
Owner CHINA UNIV OF MINING & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap