SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images
With the development of various optical sensors, change detection is one of the most actively researched areas in remotely sensed imagery with high spatial resolution. In particular, deep learning-based change detection techniques are very important for use in various fields, such as land monitoring...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2022-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9895410/ |
_version_ | 1811205885870997504 |
---|---|
author | Junsang Mo Seonkyeong Seong Jaehong Oh Jaewan Choi |
author_facet | Junsang Mo Seonkyeong Seong Jaehong Oh Jaewan Choi |
author_sort | Junsang Mo |
collection | DOAJ |
description | With the development of various optical sensors, change detection is one of the most actively researched areas in remotely sensed imagery with high spatial resolution. In particular, deep learning-based change detection techniques are very important for use in various fields, such as land monitoring and disaster analysis, because they can show superior performance compared to traditional unsupervised and supervised change detection methods. This manuscript proposes a Siamese-attentive UNet3+ for change detection (SAUNet3+CD) of multitemporal imagery with high spatial resolution. The existing UNet3+ was modified to a Siamese-based architecture, and a spatial and channel attention module was added to detect various changed areas. The proposed model was trained to effectively detect building growth and decay through the data augmentation of open datasets and a hybrid loss function. In experiments using two open datasets, the proposed deep learning model effectively detected changed areas in multitemporal images better than various methods, such as existing Siamese-based networks and a network for semantic segmentation. |
first_indexed | 2024-04-12T03:38:57Z |
format | Article |
id | doaj.art-660eaa6dadf744b7acbf28261e94026c |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-04-12T03:38:57Z |
publishDate | 2022-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-660eaa6dadf744b7acbf28261e94026c2022-12-22T03:49:21ZengIEEEIEEE Access2169-35362022-01-011010143410144410.1109/ACCESS.2022.32081349895410SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing ImagesJunsang Mo0https://orcid.org/0000-0002-3122-5734Seonkyeong Seong1Jaehong Oh2Jaewan Choi3https://orcid.org/0000-0003-3967-6481Geospatially Enabled Society Research Division, Korea Research Institute for Human Settlements, Sejong, South KoreaDepartment of Civil Engineering, Chungbuk National University, Cheongju, South KoreaDepartment of Civil Engineering, Interdisciplinary Major of Ocean Renewable Energy Engineering, Korea Maritime and Ocean University, Busan, South KoreaDepartment of Civil Engineering, Chungbuk National University, Cheongju, South KoreaWith the development of various optical sensors, change detection is one of the most actively researched areas in remotely sensed imagery with high spatial resolution. In particular, deep learning-based change detection techniques are very important for use in various fields, such as land monitoring and disaster analysis, because they can show superior performance compared to traditional unsupervised and supervised change detection methods. This manuscript proposes a Siamese-attentive UNet3+ for change detection (SAUNet3+CD) of multitemporal imagery with high spatial resolution. The existing UNet3+ was modified to a Siamese-based architecture, and a spatial and channel attention module was added to detect various changed areas. The proposed model was trained to effectively detect building growth and decay through the data augmentation of open datasets and a hybrid loss function. In experiments using two open datasets, the proposed deep learning model effectively detected changed areas in multitemporal images better than various methods, such as existing Siamese-based networks and a network for semantic segmentation.https://ieeexplore.ieee.org/document/9895410/Change detectionattention moduledeep learningUNet3+Siamese-based architecture |
spellingShingle | Junsang Mo Seonkyeong Seong Jaehong Oh Jaewan Choi SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images IEEE Access Change detection attention module deep learning UNet3+ Siamese-based architecture |
title | SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images |
title_full | SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images |
title_fullStr | SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images |
title_full_unstemmed | SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images |
title_short | SAUNet3+CD: A Siamese-Attentive UNet3+ for Change Detection in Remote Sensing Images |
title_sort | saunet3 x002b cd a siamese attentive unet3 x002b for change detection in remote sensing images |
topic | Change detection attention module deep learning UNet3+ Siamese-based architecture |
url | https://ieeexplore.ieee.org/document/9895410/ |
work_keys_str_mv | AT junsangmo saunet3x002bcdasiameseattentiveunet3x002bforchangedetectioninremotesensingimages AT seonkyeongseong saunet3x002bcdasiameseattentiveunet3x002bforchangedetectioninremotesensingimages AT jaehongoh saunet3x002bcdasiameseattentiveunet3x002bforchangedetectioninremotesensingimages AT jaewanchoi saunet3x002bcdasiameseattentiveunet3x002bforchangedetectioninremotesensingimages |