Infrared and Visible Image Fusion Based on Feature Separation

Although a pair of infrared and visible images captured in the same scene have different modes,they also have shared public information and complementary private information.A complete fusion image can be obtained by learning and integrating above information.Inspired by residual network,in the trai...

Full description

Bibliographic Details
Main Author: GAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng
Format: Article
Language:zho
Published: Editorial office of Computer Science 2022-05-01
Series:Jisuanji kexue
Subjects:
Online Access:https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-5-58.pdf
_version_ 1797845101042466816
author GAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng
author_facet GAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng
author_sort GAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng
collection DOAJ
description Although a pair of infrared and visible images captured in the same scene have different modes,they also have shared public information and complementary private information.A complete fusion image can be obtained by learning and integrating above information.Inspired by residual network,in the training stage,each branch is forced to map a label with global features through the interchange and addition of feature-levels among network branches.What’s more,each branch is encouraged to learn the private features of corresponding images.Directly learning the private features of images can avoid designing complex fusion rules and ensure the integrity of feature details.In the fusion stage,the maximum fusion strategy is adopted to fuse the private features,add them to the learned public features at the decoding layer and finally decode the fused image.The model is trained over a multi-focused data set that is synthesized from the NYU-D2 and tested over the real-world TNO data set.Experimental results show that compared with the current mainstream infrared and visible fusion algorithms,the proposed algorithm achieves better results in subjective effects and objective evaluation indicators.
first_indexed 2024-04-09T17:33:05Z
format Article
id doaj.art-f55ada0408114053833c4c0995b547d3
institution Directory Open Access Journal
issn 1002-137X
language zho
last_indexed 2024-04-09T17:33:05Z
publishDate 2022-05-01
publisher Editorial office of Computer Science
record_format Article
series Jisuanji kexue
spelling doaj.art-f55ada0408114053833c4c0995b547d32023-04-18T02:35:57ZzhoEditorial office of Computer ScienceJisuanji kexue1002-137X2022-05-01495586310.11896/jsjkx.210200148Infrared and Visible Image Fusion Based on Feature SeparationGAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng01 School of Artificial Intelligence and Computer,Jiangsu University,Wuxi,Jiangsu 214122,China ;2 Pattern Recognition and Computational Intelligence Engineering Laboratory,Wuxi,Jiangsu 214122,China ;3 School of Electronics and Information Engineering,Suzhou University of Science and Technology,Suzhou,Jiangsu 215009,ChinaAlthough a pair of infrared and visible images captured in the same scene have different modes,they also have shared public information and complementary private information.A complete fusion image can be obtained by learning and integrating above information.Inspired by residual network,in the training stage,each branch is forced to map a label with global features through the interchange and addition of feature-levels among network branches.What’s more,each branch is encouraged to learn the private features of corresponding images.Directly learning the private features of images can avoid designing complex fusion rules and ensure the integrity of feature details.In the fusion stage,the maximum fusion strategy is adopted to fuse the private features,add them to the learned public features at the decoding layer and finally decode the fused image.The model is trained over a multi-focused data set that is synthesized from the NYU-D2 and tested over the real-world TNO data set.Experimental results show that compared with the current mainstream infrared and visible fusion algorithms,the proposed algorithm achieves better results in subjective effects and objective evaluation indicators.https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-5-58.pdfresidual learning|feature extraction|private feature||public feature|image fusion
spellingShingle GAO Yuan-hao, LUO Xiao-qing, ZHANG Zhan-cheng
Infrared and Visible Image Fusion Based on Feature Separation
Jisuanji kexue
residual learning|feature extraction|private feature||public feature|image fusion
title Infrared and Visible Image Fusion Based on Feature Separation
title_full Infrared and Visible Image Fusion Based on Feature Separation
title_fullStr Infrared and Visible Image Fusion Based on Feature Separation
title_full_unstemmed Infrared and Visible Image Fusion Based on Feature Separation
title_short Infrared and Visible Image Fusion Based on Feature Separation
title_sort infrared and visible image fusion based on feature separation
topic residual learning|feature extraction|private feature||public feature|image fusion
url https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-5-58.pdf
work_keys_str_mv AT gaoyuanhaoluoxiaoqingzhangzhancheng infraredandvisibleimagefusionbasedonfeatureseparation