Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection
Change detection is an important task in Earth observation with applications in environmental monitoring, urban development, and disaster management. Traditional supervised deep learning approaches rely on labeled bitemporal datasets, which are often scarce, making unsupervised methods, such as deep...
Saved in:
| Main Authors: | , |
|---|---|
| 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/11086368/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849391083236098048 |
|---|---|
| author | Sudipan Saha Kanishk Awadhiya |
| author_facet | Sudipan Saha Kanishk Awadhiya |
| author_sort | Sudipan Saha |
| collection | DOAJ |
| description | Change detection is an important task in Earth observation with applications in environmental monitoring, urban development, and disaster management. Traditional supervised deep learning approaches rely on labeled bitemporal datasets, which are often scarce, making unsupervised methods, such as deep change vector analysis (DCVA), a popular choice. However, unsupervised methods detect generic changes without distinguishing between different types, limiting their applicability in scenarios where class-specific change detection is required. Recently, foundation models, such as the segment anything model (SAM), have demonstrated strong generalization capabilities, allowing for precise semantic segmentation with minimal supervision. Leveraging these advancements, we propose a novel framework that integrates DCVA for unsupervised change detection with SAM’s ability to segment specific targets using only a few examples. This hybrid approach enables low-cost, class-specific change detection, reducing the need for extensive labeled datasets while improving the interpretability and relevance of detected changes. The proposed method holds significant potential for targeted monitoring applications in Earth observation. We demonstrate its capability for three classes related to buildings, trees, and landslide. |
| format | Article |
| id | doaj-art-4135295c3bfb47069d7c00ab7eb35d5c |
| institution | Kabale University |
| 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-4135295c3bfb47069d7c00ab7eb35d5c2025-08-20T03:41:11ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118184391844910.1109/JSTARS.2025.359101711086368Integrating Deep Change Vector Analysis and SAM for Class-Specific Change DetectionSudipan Saha0https://orcid.org/0000-0002-9440-0720Kanishk Awadhiya1https://orcid.org/0009-0006-0557-0936Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi, New Delhi, IndiaYardi School of Artificial Intelligence, Indian Institute of Technology Delhi, New Delhi, IndiaChange detection is an important task in Earth observation with applications in environmental monitoring, urban development, and disaster management. Traditional supervised deep learning approaches rely on labeled bitemporal datasets, which are often scarce, making unsupervised methods, such as deep change vector analysis (DCVA), a popular choice. However, unsupervised methods detect generic changes without distinguishing between different types, limiting their applicability in scenarios where class-specific change detection is required. Recently, foundation models, such as the segment anything model (SAM), have demonstrated strong generalization capabilities, allowing for precise semantic segmentation with minimal supervision. Leveraging these advancements, we propose a novel framework that integrates DCVA for unsupervised change detection with SAM’s ability to segment specific targets using only a few examples. This hybrid approach enables low-cost, class-specific change detection, reducing the need for extensive labeled datasets while improving the interpretability and relevance of detected changes. The proposed method holds significant potential for targeted monitoring applications in Earth observation. We demonstrate its capability for three classes related to buildings, trees, and landslide.https://ieeexplore.ieee.org/document/11086368/Change detection (CD)deep learningEarth observationfoundation models |
| spellingShingle | Sudipan Saha Kanishk Awadhiya Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Change detection (CD) deep learning Earth observation foundation models |
| title | Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection |
| title_full | Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection |
| title_fullStr | Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection |
| title_full_unstemmed | Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection |
| title_short | Integrating Deep Change Vector Analysis and SAM for Class-Specific Change Detection |
| title_sort | integrating deep change vector analysis and sam for class specific change detection |
| topic | Change detection (CD) deep learning Earth observation foundation models |
| url | https://ieeexplore.ieee.org/document/11086368/ |
| work_keys_str_mv | AT sudipansaha integratingdeepchangevectoranalysisandsamforclassspecificchangedetection AT kanishkawadhiya integratingdeepchangevectoranalysisandsamforclassspecificchangedetection |