Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images
Satellite image change detection, where two images of the same area from different times are compared, is crucial for earth sensing and monitoring applications. Many learning-based detection methods have been proposed for this task, with different performance characteristics. Since these detection m...
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/10815622/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590370925445120 |
|---|---|
| author | Maria-Eirini Pegia Bjorn or Jonsson Anastasia Moumtzidou Ilias Gialampoukidis Stefanos Vrochidis Ioannis Kompatsiaris |
| author_facet | Maria-Eirini Pegia Bjorn or Jonsson Anastasia Moumtzidou Ilias Gialampoukidis Stefanos Vrochidis Ioannis Kompatsiaris |
| author_sort | Maria-Eirini Pegia |
| collection | DOAJ |
| description | Satellite image change detection, where two images of the same area from different times are compared, is crucial for earth sensing and monitoring applications. Many learning-based detection methods have been proposed for this task, with different performance characteristics. Since these detection methods have been tested under different settings, comparing their performance across a variety of situations is difficult. The goal of this article is therefore to comprehensively compare the state-of-the-art detection methods from the literature, across a variety of dataset parameters. To that end, we analyze the impact of image resolution, training set size, and noise on learning performance. A first set of experiments, using a large set of high-resolution images, reveals that training set resolution should match the resolution of the images the model will be applied to, that larger training sets are beneficial, and that adding Gaussian noise improves performance. A second set of experiments, using a smaller set of low-resolution images, confirms that the training set should also be of the same low resolution, but shows that adding noise does not improve performance in this case. The results also indicate that BiasUNet is the most effective method for detecting changes between image pairs. |
| format | Article |
| id | doaj-art-0308883188af4b1f82c3ce2a38487e93 |
| 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-0308883188af4b1f82c3ce2a38487e932025-01-24T00:01:01ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01183766378110.1109/JSTARS.2024.352235010815622Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite ImagesMaria-Eirini Pegia0https://orcid.org/0000-0003-2643-0028Bjorn or Jonsson1https://orcid.org/0000-0003-0889-3491Anastasia Moumtzidou2https://orcid.org/0000-0001-7615-8400Ilias Gialampoukidis3https://orcid.org/0000-0002-5234-9795Stefanos Vrochidis4https://orcid.org/0000-0002-2505-9178Ioannis Kompatsiaris5https://orcid.org/0000-0001-6447-9020Technologies Institute, Centre for Research and Technology, Hellas, GreeceTechnologies Institute, Centre for Research and Technology, Hellas, GreeceTechnologies Institute, Centre for Research and Technology, Hellas, GreeceTechnologies Institute, Centre for Research and Technology, Hellas, GreeceTechnologies Institute, Centre for Research and Technology, Hellas, GreeceTechnologies Institute, Centre for Research and Technology, Hellas, GreeceSatellite image change detection, where two images of the same area from different times are compared, is crucial for earth sensing and monitoring applications. Many learning-based detection methods have been proposed for this task, with different performance characteristics. Since these detection methods have been tested under different settings, comparing their performance across a variety of situations is difficult. The goal of this article is therefore to comprehensively compare the state-of-the-art detection methods from the literature, across a variety of dataset parameters. To that end, we analyze the impact of image resolution, training set size, and noise on learning performance. A first set of experiments, using a large set of high-resolution images, reveals that training set resolution should match the resolution of the images the model will be applied to, that larger training sets are beneficial, and that adding Gaussian noise improves performance. A second set of experiments, using a smaller set of low-resolution images, confirms that the training set should also be of the same low resolution, but shows that adding noise does not improve performance in this case. The results also indicate that BiasUNet is the most effective method for detecting changes between image pairs.https://ieeexplore.ieee.org/document/10815622/Change detectiondeep learningsatellite data |
| spellingShingle | Maria-Eirini Pegia Bjorn or Jonsson Anastasia Moumtzidou Ilias Gialampoukidis Stefanos Vrochidis Ioannis Kompatsiaris Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Change detection deep learning satellite data |
| title | Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images |
| title_full | Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images |
| title_fullStr | Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images |
| title_full_unstemmed | Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images |
| title_short | Comparative Analysis of Learning-Based Approaches for Change Detection in Satellite Images |
| title_sort | comparative analysis of learning based approaches for change detection in satellite images |
| topic | Change detection deep learning satellite data |
| url | https://ieeexplore.ieee.org/document/10815622/ |
| work_keys_str_mv | AT mariaeirinipegia comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages AT bjornorjonsson comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages AT anastasiamoumtzidou comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages AT iliasgialampoukidis comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages AT stefanosvrochidis comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages AT ioanniskompatsiaris comparativeanalysisoflearningbasedapproachesforchangedetectioninsatelliteimages |