Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection
Change detection (CD) with heterogeneous images is currently attracting extensive attention in remote sensing. In order to make heterogeneous images comparable, the image transformation methods transform one image into the domain of another image, which can simultaneously obtain a forward difference...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/21/5622 |
| _version_ | 1797466500605411328 |
|---|---|
| author | Kuowei Xiao Yuli Sun Lin Lei |
| author_facet | Kuowei Xiao Yuli Sun Lin Lei |
| author_sort | Kuowei Xiao |
| collection | DOAJ |
| description | Change detection (CD) with heterogeneous images is currently attracting extensive attention in remote sensing. In order to make heterogeneous images comparable, the image transformation methods transform one image into the domain of another image, which can simultaneously obtain a forward difference map (FDM) and backward difference map (BDM). However, previous methods only fuse the FDM and BDM in the post-processing stage, which cannot fundamentally improve the performance of CD. In this paper, a change alignment-based change detection (CACD) framework for unsupervised heterogeneous CD is proposed to deeply utilize the complementary information of the FDM and BDM in the image transformation process, which enhances the effect of domain transformation, thus improving CD performance. To reduce the dependence of the transformation network on labeled samples, we propose a graph structure-based strategy of generating prior masks to guide the network, which can reduce the influence of changing regions on the transformation network in an unsupervised way. More importantly, based on the fact that the FDM and BDM are representing the same change event, we perform change alignment during the image transformation, which can enhance the image transformation effect and enable FDM and BDM to effectively indicate the real change region. Comparative experiments are conducted with six state-of-the-art methods on five heterogeneous CD datasets, showing that the proposed CACD achieves the best performance with an average overall accuracy (OA) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>95.9</mn><mo>%</mo></mrow></semantics></math></inline-formula> on different datasets and at least <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>6.8</mn><mo>%</mo></mrow></semantics></math></inline-formula> improvement in the kappa coefficient. |
| first_indexed | 2024-03-09T18:40:41Z |
| format | Article |
| id | doaj.art-01965ad3ee7444019a716631823cac52 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T18:40:41Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-01965ad3ee7444019a716631823cac522023-11-24T06:42:01ZengMDPI AGRemote Sensing2072-42922022-11-011421562210.3390/rs14215622Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change DetectionKuowei Xiao0Yuli Sun1Lin Lei2College of Electronic Science, National University of Defense Technology, Changsha 410073, ChinaCollege of Electronic Science, National University of Defense Technology, Changsha 410073, ChinaCollege of Electronic Science, National University of Defense Technology, Changsha 410073, ChinaChange detection (CD) with heterogeneous images is currently attracting extensive attention in remote sensing. In order to make heterogeneous images comparable, the image transformation methods transform one image into the domain of another image, which can simultaneously obtain a forward difference map (FDM) and backward difference map (BDM). However, previous methods only fuse the FDM and BDM in the post-processing stage, which cannot fundamentally improve the performance of CD. In this paper, a change alignment-based change detection (CACD) framework for unsupervised heterogeneous CD is proposed to deeply utilize the complementary information of the FDM and BDM in the image transformation process, which enhances the effect of domain transformation, thus improving CD performance. To reduce the dependence of the transformation network on labeled samples, we propose a graph structure-based strategy of generating prior masks to guide the network, which can reduce the influence of changing regions on the transformation network in an unsupervised way. More importantly, based on the fact that the FDM and BDM are representing the same change event, we perform change alignment during the image transformation, which can enhance the image transformation effect and enable FDM and BDM to effectively indicate the real change region. Comparative experiments are conducted with six state-of-the-art methods on five heterogeneous CD datasets, showing that the proposed CACD achieves the best performance with an average overall accuracy (OA) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>95.9</mn><mo>%</mo></mrow></semantics></math></inline-formula> on different datasets and at least <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>6.8</mn><mo>%</mo></mrow></semantics></math></inline-formula> improvement in the kappa coefficient.https://www.mdpi.com/2072-4292/14/21/5622change detectionheterogeneous imageunsupervisedchange alignmentimage transformationprior mask |
| spellingShingle | Kuowei Xiao Yuli Sun Lin Lei Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection Remote Sensing change detection heterogeneous image unsupervised change alignment image transformation prior mask |
| title | Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection |
| title_full | Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection |
| title_fullStr | Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection |
| title_full_unstemmed | Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection |
| title_short | Change Alignment-Based Image Transformation for Unsupervised Heterogeneous Change Detection |
| title_sort | change alignment based image transformation for unsupervised heterogeneous change detection |
| topic | change detection heterogeneous image unsupervised change alignment image transformation prior mask |
| url | https://www.mdpi.com/2072-4292/14/21/5622 |
| work_keys_str_mv | AT kuoweixiao changealignmentbasedimagetransformationforunsupervisedheterogeneouschangedetection AT yulisun changealignmentbasedimagetransformationforunsupervisedheterogeneouschangedetection AT linlei changealignmentbasedimagetransformationforunsupervisedheterogeneouschangedetection |