ABSTRACT
Change detection (CD) is an important and difficult task in remote sensing image interpretation. Most of the existing deep learning-based methods try to design complex networks to improve the accuracy of detection. However, most of these methods ignore pseudo-changes caused by temporal changes, such as seasonal differences and illumination fluctuations between bi-temporal images, resulting in sub-optimal results. In this paper, we propose a Retinex decomposition module to alleviate the influence of factors such as illumination on CD accuracy by Retinex decomposition. In addition, we design a simple Unet-like CD module for CD of the reflectance maps obtained by Retinex decomposition. Since the Retinex decomposition module has eliminated most of the impact of pseudo-changes on the CD results, the proposed Retinex decomposition-based change detection network (R-CDNet) can achieve better results. The results on the WHU-CD dataset demonstrate that the proposed R-CDNet has better overall accuracy.
- Qiqi Zhu, Land-use/land-cover change detection based on a Siamese global learning framework for high spatial resolution remote sensing imagery. ISPRS Journal of Photogrammetry and Remote Sensing 184 (2022): 63-78.Google ScholarCross Ref
- Feng Gao, Xiao Wang, Yunhao Gao, Junyu Dong, and Shengke Wang, Sea Ice Change Detection in SAR Images Based on Convolutional-Wavelet Neural Networks, in IEEE Geoscience and Remote Sensing Letters, vol.16, no.8, pp.1240-1244, Aug.2019, doi: 10.1109/LGRS.2019.2895656.Google ScholarCross Ref
- Huijiao Qiao, A novel change detection method for natural disaster detection and segmentation from video sequence. Sensors , (2020), 18: 5076.Google Scholar
- Zhi Zheng, CLNet: Cross-layer convolutional neural network for change detection in optical remote sensing imagery. ISPRS Journal of Photogrammetry and Remote Sensing 175 (2021): 247-267.Google ScholarCross Ref
- Zhi Zheng, CLNet: Cross-layer convolutional neural network for change detection in optical remote sensing imagery. ISPRS Journal of Photogrammetry and Remote Sensing 175 (2021): 247-267.Google ScholarCross Ref
- Min Zhang, and Wenzhong Shi. A feature difference convolutional neural network-based change detection method. IEEE Transactions on Geoscience and Remote Sensing 58.10 (2020): 7232-7246.Google ScholarCross Ref
- Guizhou Wang, Guojin He, Jianbo Liu, Mengmeng Wang, and Bo Cheng, "Land Cover Change Monitoring of Mine City Using Multi-Temporal Satellite Remote Sensing Images," Vol. 8, No. 1, pp. 17-22, February, 2017. doi: 10.12720/jait.8.1.17-22.Google ScholarCross Ref
- Tao Lei, , Difference Enhancement and Spatial–Spectral Nonlocal Network for Change Detection in VHR Remote Sensing Images," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-13, 2022, Art no. 4507013, doi: 10.1109/TGRS.2021.3134691.Google ScholarCross Ref
- Wele G. C. Bandara and Vishal M. Patel, A Transformer-Based Siamese Network for Change Detection, IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia, 2022, pp. 207-210, doi: 10.1109/IGARSS46834.2022.9883686.Google ScholarCross Ref
- Qingyang Li, Ruofei Zhong, Xin Du, and Yu Du, TransUNetCD: A Hybrid Transformer Network for Change Detection in Optical Remote-Sensing Images," in IEEE Transactions on Geoscience and Remote Sensing, vol. 60 (2022), pp. 1-19, Art no. 5622519, doi 10.1109/TGRS.2022.3169479.Google Scholar
- Tianyu Yan, Zifu Wan, and Pingping Zhang. Fully transformer network for change detection of remote sensing images. Proceedings of the Asian Conference on Computer Vision. 2022.Google Scholar
- Shike Liang, Zhen Hua, and Jinjiang Li. GCN-based multi-scale dual fusion for remote sensing building change detection. International Journal of Remote Sensing 44.3 (2023): 953-980.Google ScholarCross Ref
- Sudipan Saha, Semisupervised change detection using graph convolutional network. IEEE Geoscience and Remote Sensing Letters 18.4 (2020): 607-611.Google ScholarCross Ref
- Jia Liu, An end-to-end supervised domain adaptation framework for cross-domain change detection. Pattern Recognition 132 (2022): 108960.Google ScholarDigital Library
- Edwin H. Land. The Retinex Theory of Color Vision. Scientific American 237, no. 6 (1977): 108–29. http://www.jstor.org/stable/24953876.Google ScholarCross Ref
- Risheng Liu, Retinex-inspired unrolling with cooperative prior architecture search for low-light image enhancement. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021.Google ScholarCross Ref
- Peixian Zhuang, Chongyi Li, and Jiamin Wu. Bayesian retinex underwater image enhancement. Engineering Applications of Artificial Intelligence 101 (2021): 104171.Google ScholarCross Ref
- Wenhan Yang, Sparse gradient regularized deep retinex network for robust low-light image enhancement. IEEE Transactions on Image Processing 30 (2021): 2072-2086.Google ScholarCross Ref
- Shunping Ji, Shiqing Wei, Meng Lu, Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set. IEEE Transactions on Geoscience and Remote Sensing vol. 57 (2018),pp 574–586.Google ScholarCross Ref
- Rodrigo C. Daudt, Bertr Le Saux, and Alexandre Boulch, Fully convolutional Siamese networks for change detection. IEEE International Conference on Image Processing (2018) 4063–4067.Google Scholar
Index Terms
- Retinex Decomposition-based Change Detection Network for HR Remote Sensing Images
Recommendations
End-to-End Remote Sensing Change Detection of Unregistered Bi-temporal Images for Natural Disasters
Artificial Neural Networks and Machine Learning – ICANN 2023AbstractChange detection based on remote sensing images has been a prominent area of interest in the field of remote sensing. Deep networks have demonstrated significant success in detecting changes in bi-temporal remote sensing images and have found ...
Semi-supervised Zero-Shot Learning in Retinex decomposition-based Low-Light Image Enhancement
ICCAI '23: Proceedings of the 2023 9th International Conference on Computing and Artificial IntelligenceThe images recorded by people are often subject to multiple degradation challenges due to the shooting environment and shooting equipment. In low-light scenes, images acquired by non-professional equipment are usually accompanied by a large amount of ...
Remote Sensing Image Enhancement Method Based on Multi-scale Retinex
ICM '11: Proceedings of the 2011 International Conference of Information Technology, Computer Engineering and Management Sciences - Volume 03In order to improve the effectiveness of remote sensing image enhancement, an enhancement algorithm based on multi-scale Retinex was presented and which can make up the deficiency of traditional wavelet algorithm that losing part information when image ...
Comments