SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks

The goal of image fusion is to retain the strengths of different images in the fused result. However, existing fusion algorithms are often complex in design and overlook the influence of attention mechanisms on deep features. To address these issues, we propose an image fusion network based on spati...

Full description

Bibliographic Details
Main Authors: Wang, Yong, Pu, Jianfei, Miao, Duoqian, Zhang, Longbin, Zhang, Lulu, Du, Xin
Other Authors: School of Mechanical and Aerospace Engineering
Format: Journal Article
Language:English
Published: 2024
Subjects:
Online Access:https://hdl.handle.net/10356/180177
_version_ 1826110230771007488
author Wang, Yong
Pu, Jianfei
Miao, Duoqian
Zhang, Longbin
Zhang, Lulu
Du, Xin
author2 School of Mechanical and Aerospace Engineering
author_facet School of Mechanical and Aerospace Engineering
Wang, Yong
Pu, Jianfei
Miao, Duoqian
Zhang, Longbin
Zhang, Lulu
Du, Xin
author_sort Wang, Yong
collection NTU
description The goal of image fusion is to retain the strengths of different images in the fused result. However, existing fusion algorithms are often complex in design and overlook the influence of attention mechanisms on deep features. To address these issues, we propose an image fusion network based on spatial/channel attention mechanisms and gradient-aggregated residual dense blocks(SCGRFuse). Firstly, we design a novel gradient-aggregated residual dense block (GRXDB) that combines the advantages of ResNeXt and DenseNet, which integrating the Sobel and Laplacian operators to preserve both strong and weak texture features. Then, we introduce spatial and channel attention mechanisms to refine the channel and spatial information of feature maps, enhancing their information capturing capability. Additionally, we leverage a pooling fusion block to merge the refined spatial and channel feature maps, yielding high-quality fusion features. Compared to the existing state-of-the-art methods, experimental results on the MSRS, RoadScene and TNO datasets demonstrate the outstanding fusion performance of our proposed approach. In addition, in the task-driven experiments, SCGRFuse achieved an mIoU accuracy of 71.37%.
first_indexed 2024-10-01T02:31:01Z
format Journal Article
id ntu-10356/180177
institution Nanyang Technological University
language English
last_indexed 2024-10-01T02:31:01Z
publishDate 2024
record_format dspace
spelling ntu-10356/1801772024-09-23T04:35:40Z SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks Wang, Yong Pu, Jianfei Miao, Duoqian Zhang, Longbin Zhang, Lulu Du, Xin School of Mechanical and Aerospace Engineering Rehabilitation Research Institute of Singapore (RRIS) Engineering Image fusion Infrared image The goal of image fusion is to retain the strengths of different images in the fused result. However, existing fusion algorithms are often complex in design and overlook the influence of attention mechanisms on deep features. To address these issues, we propose an image fusion network based on spatial/channel attention mechanisms and gradient-aggregated residual dense blocks(SCGRFuse). Firstly, we design a novel gradient-aggregated residual dense block (GRXDB) that combines the advantages of ResNeXt and DenseNet, which integrating the Sobel and Laplacian operators to preserve both strong and weak texture features. Then, we introduce spatial and channel attention mechanisms to refine the channel and spatial information of feature maps, enhancing their information capturing capability. Additionally, we leverage a pooling fusion block to merge the refined spatial and channel feature maps, yielding high-quality fusion features. Compared to the existing state-of-the-art methods, experimental results on the MSRS, RoadScene and TNO datasets demonstrate the outstanding fusion performance of our proposed approach. In addition, in the task-driven experiments, SCGRFuse achieved an mIoU accuracy of 71.37%. This work is supported by the National Natural Science Foundation of China under Grant 61976158 and Grant 61673301. 2024-09-23T04:35:40Z 2024-09-23T04:35:40Z 2024 Journal Article Wang, Y., Pu, J., Miao, D., Zhang, L., Zhang, L. & Du, X. (2024). SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks. Engineering Applications of Artificial Intelligence, 132, 107898-. https://dx.doi.org/10.1016/j.engappai.2024.107898 0952-1976 https://hdl.handle.net/10356/180177 10.1016/j.engappai.2024.107898 2-s2.0-85184995954 132 107898 en Engineering Applications of Artificial Intelligence © 2024 Elsevier Ltd. All rights reserved.
spellingShingle Engineering
Image fusion
Infrared image
Wang, Yong
Pu, Jianfei
Miao, Duoqian
Zhang, Longbin
Zhang, Lulu
Du, Xin
SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title_full SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title_fullStr SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title_full_unstemmed SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title_short SCGRFuse: an infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks
title_sort scgrfuse an infrared and visible image fusion network based on spatial channel attention mechanism and gradient aggregation residual dense blocks
topic Engineering
Image fusion
Infrared image
url https://hdl.handle.net/10356/180177
work_keys_str_mv AT wangyong scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks
AT pujianfei scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks
AT miaoduoqian scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks
AT zhanglongbin scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks
AT zhanglulu scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks
AT duxin scgrfuseaninfraredandvisibleimagefusionnetworkbasedonspatialchannelattentionmechanismandgradientaggregationresidualdenseblocks