A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images

Accurate disparity estimation of high-resolution satellite remote sensing stereo images serves as a crucial method for generating precise digital surface models. However, the complex intractable regions in satellite images (textureless regions, repeated texture regions, occlusion regions) pose serio...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhenghui Xu, Yonghua Jiang, Jingxue Wang, Yunming Wang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10758818/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850164935817953280
author Zhenghui Xu
Yonghua Jiang
Jingxue Wang
Yunming Wang
author_facet Zhenghui Xu
Yonghua Jiang
Jingxue Wang
Yunming Wang
author_sort Zhenghui Xu
collection DOAJ
description Accurate disparity estimation of high-resolution satellite remote sensing stereo images serves as a crucial method for generating precise digital surface models. However, the complex intractable regions in satellite images (textureless regions, repeated texture regions, occlusion regions) pose serious challenges for accurate disparity estimation. To enhance the matching accuracy within intractable regions, a dual branch multiscale stereo matching network for high-resolution satellite stereo images is proposed. First, a dual branch feature extraction module is designed, which can perform efficient downsampling. This module can enhance the scene awareness capability of the model, enabling it to extract multiscale feature maps and construct multiscale cost volumes. Then, the cost aggregation process is executed in a coarse-to-fine manner. The method employs a simple hourglass structure and leverages low-scale information to guide the aggregation of high-scale cost volumes. Next, a disparity-channel attention mechanism is proposed for the cost aggregation process to obtain more representative feature information. Finally, a simple disparity refinement module is designed by utilizing both intensity and gradient information of the left image to improve the local details of the disparity map. Experiments are performed separately on the GaoFen-7 and US3D datasets. The experimental results indicate that the proposed method is conducive to improving the matching accuracy within intractable regions of satellite images. The structure of the proposed network is simple, which can effectively reduce the network parameters and realize the lightweight of the model.
format Article
id doaj-art-41646e4d795f4935b533b17f0faf0d56
institution OA Journals
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-41646e4d795f4935b533b17f0faf0d562025-08-20T02:21:51ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-011894996410.1109/JSTARS.2024.350284210758818A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing ImagesZhenghui Xu0https://orcid.org/0000-0003-3550-8720Yonghua Jiang1https://orcid.org/0000-0001-5777-4144Jingxue Wang2https://orcid.org/0000-0002-8816-4362Yunming Wang3School of Geomatics, Liaoning Technical University, Fuxin, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Automation and Electronic Information, Xiangtan University, Xiangtan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaAccurate disparity estimation of high-resolution satellite remote sensing stereo images serves as a crucial method for generating precise digital surface models. However, the complex intractable regions in satellite images (textureless regions, repeated texture regions, occlusion regions) pose serious challenges for accurate disparity estimation. To enhance the matching accuracy within intractable regions, a dual branch multiscale stereo matching network for high-resolution satellite stereo images is proposed. First, a dual branch feature extraction module is designed, which can perform efficient downsampling. This module can enhance the scene awareness capability of the model, enabling it to extract multiscale feature maps and construct multiscale cost volumes. Then, the cost aggregation process is executed in a coarse-to-fine manner. The method employs a simple hourglass structure and leverages low-scale information to guide the aggregation of high-scale cost volumes. Next, a disparity-channel attention mechanism is proposed for the cost aggregation process to obtain more representative feature information. Finally, a simple disparity refinement module is designed by utilizing both intensity and gradient information of the left image to improve the local details of the disparity map. Experiments are performed separately on the GaoFen-7 and US3D datasets. The experimental results indicate that the proposed method is conducive to improving the matching accuracy within intractable regions of satellite images. The structure of the proposed network is simple, which can effectively reduce the network parameters and realize the lightweight of the model.https://ieeexplore.ieee.org/document/10758818/Disparity-channel attentiondual branch feature extraction (DBFE)satellite remote sensing stereo imagesstereo matching
spellingShingle Zhenghui Xu
Yonghua Jiang
Jingxue Wang
Yunming Wang
A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disparity-channel attention
dual branch feature extraction (DBFE)
satellite remote sensing stereo images
stereo matching
title A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
title_full A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
title_fullStr A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
title_full_unstemmed A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
title_short A Dual Branch Multiscale Stereo Matching Network for High-Resolution Satellite Remote Sensing Images
title_sort dual branch multiscale stereo matching network for high resolution satellite remote sensing images
topic Disparity-channel attention
dual branch feature extraction (DBFE)
satellite remote sensing stereo images
stereo matching
url https://ieeexplore.ieee.org/document/10758818/
work_keys_str_mv AT zhenghuixu adualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT yonghuajiang adualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT jingxuewang adualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT yunmingwang adualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT zhenghuixu dualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT yonghuajiang dualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT jingxuewang dualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages
AT yunmingwang dualbranchmultiscalestereomatchingnetworkforhighresolutionsatelliteremotesensingimages