ABSTRACT
Deep Convolutional Neural Networks have been adopted for remote sensing change detection that focused on how to migrate semantic segmentation networks designed for a single image to remote sensing change detection tasks. These networks tend to have an accuracy of the large region rather than boundary quality and small region quality. In this paper, we propose a novel architecture, Siamese Change Detection Network (SCD-Net), and a new hybrid loss, Multi-Scale Perceptual (MSP) Loss, for bi-temporal remote sensing change detection. Specifically, the architecture is composed of a densely supervised Encoder-Decoder network in which, unlike existing work, we add an up-sampling path to the encoder in charge of building multi-level strong semantic feature maps. In this way, the comparison of low-level feature maps is based on global information prior instead of only local information. The Multi-Scale Perceptual (MSP) Loss consists of Tversky loss and a variant of Focal loss. It is applied to the output results of the network at different scales to be able to learn the changed regions at different scales effectively. Equipped with MSP loss, the proposed SCD-Net can effectively segment the change regions and accurately predict the fine structures with accurate boundaries. Experimental results on two public datasets show that our method outperforms the state-of-the-art methods in F1 score.
- Ashbindu Singh. 1989. Review article digital change detection techniques using remotely-sensed data. International journal of remote sensing 10, 6 (1989), 989–1003.Google ScholarCross Ref
- Tobias Leichtle, Christian Geiß, Michael Wurm, Tobia Lakes, and Hannes Taubenböck. 2017. Unsupervised change detection in VHR remote sensing imagery–an object-based clustering approach in a dynamic urban environment. International Journal of Applied Earth Observation and Geoinformation 54 (2017), 15–27.Google ScholarCross Ref
- Dengsheng Lu, Paul Mausel, Eduardo Brondizio, and Emilio Moran. 2004. Change detection techniques. International journal of remote sensing 25, 12 (2004), 2365–2401.Google ScholarCross Ref
- Wenzhong Shi, Min Zhang, Hongfei Ke, Xin Fang, Zhao Zhan, and Shanxiong Chen. 2020. Landslide recognition by deep convolutional neural network and change detection. IEEE Transactions on Geoscience and Remote Sensing 59, 6 (2020), 4654–4672. DOI: 10.1109/TGRS.2020.3015826Google ScholarCross Ref
- Xin Huang, Yinxia Cao, and Jiayi Li. 2020. An automatic change detection method for monitoring newly constructed building areas using time-series multi-view high-resolution optical satellite images. Remote Sensing of Environment 244 (2020), 111802.Google ScholarCross Ref
- Jonathan Long, Evan Shelhamer, and Trevor Darrell. 2015. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. 3431–3440.Google ScholarCross Ref
- Karen Simonyan and Andrew Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014).Google Scholar
- Rodrigo Caye Daudt, Bertr Le Saux, Alexandre Boulch, and Yann Gousseau. 2018. Urban change detection for multispectral earth observation using convolutional neural networks. In IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2115–2118.Google ScholarCross Ref
- Ying Chen, Xu Ouyang, and Gady Agam. 2019. ChangeNet: Learning to detect changes in satellite images. In Proceedings of the 3rd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery. 24–31.Google ScholarDigital Library
- MA Lebedev, Yu V Vizilter, OV Vygolov, VA Knyaz, and A Yu Rubis. 2018. CHANGE DETECTION IN REMOTE SENSING IMAGES USING CONDITIONAL ADVERSARIAL NETWORKS. International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences 42, 2 (2018).Google ScholarCross Ref
- Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention. Springer, 234–241.Google ScholarCross Ref
- Yanan You, Jingyi Cao, and Wenli Zhou. 2020. A survey of change detection methods based on remote sensing images for multi-source and multi-objective scenarios. Remote Sensing 12, 15 (2020), 2460.Google ScholarCross Ref
- Rodrigo Caye Daudt, Bertr Le Saux, and Alexandre Boulch. 2018. Fully convolutional siamese networks for change detection. In 2018 25th IEEE International Conference on Image Processing (ICIP). IEEE, 4063–4067.Google ScholarCross Ref
- Seyed Sadegh Mohseni Salehi, Deniz Erdogmus, and Ali Gholipour. 2017. Tversky loss function for image segmentation using 3D fully convolutional deep networks. In International workshop on machine learning in medical imaging. Springer, 379–387.Google ScholarCross Ref
- Masroor Hussain, Dongmei Chen, Angela Cheng, Hui Wei, and David Stanley. 2013. Change detection from remotely sensed images: From pixel-based to object-based approaches. ISPRS Journal of photogrammetry and remote sensing 80 (2013), 91–106.Google Scholar
- Maoguo Gong, Hailun Yang, and Puzhao Zhang. 2017. Feature learning and change feature classification based on deep learning for ternary change detection in SAR images. ISPRS Journal of Photogrammetry and Remote Sensing 129 (2017), 212–225.Google ScholarCross Ref
- Lichao Mou, Lorenzo Bruzzone, and Xiao Xiang Zhu. 2018. Learning spectral-spatial-temporal features via a recurrent convolutional neural network for change detection in multispectral imagery. IEEE Transactions on Geoscience and Remote Sensing 57, 2 (2018), 924–935. DOI: 10.1109/TGRS.2018.2863224Google ScholarCross Ref
- Yang Zhan, Kun Fu, Menglong Yan, Xian Sun, Hongqi Wang, and Xiaosong Qiu. 2017. Change detection based on deep siamese convolutional network for optical aerial images. IEEE Geoscience and Remote Sensing Letters 14, 10 (2017), 1845–1849. DOI: 10.1109/LGRS.2017.2738149Google ScholarCross Ref
- Qing Wang, Xiaodong Zhang, Guanzhou Chen, Fan Dai, Yuanfu Gong, and Kun Zhu. 2018. Change detection based on Faster R-CNN for high-resolution remote sensing images. Remote sensing letters 9, 10 (2018), 923–932.Google Scholar
- Meijuan Yang, Licheng Jiao, Fang Liu, Biao Hou, and Shuyuan Yang. 2019. Transferred deep learning-based change detection in remote sensing images. IEEE Transactions on Geoscience and Remote Sensing 57, 9 (2019), 6960–6973. DOI: 10.1109/TGRS.2019.2909781Google ScholarCross Ref
- Daifeng Peng, Yongjun Zhang, and Haiyan Guan. 2019. End-to-end change detection for high resolution satellite images using improved UNet++. Remote Sensing 11, 11 (2019), 1382.Google ScholarCross Ref
- Chenxiao Zhang, Peng Yue, Deodato Tapete, Liangcun Jiang, Boyi Shangguan, Li Huang, and Guangchao Liu. 2020. A deeply supervised image fusion network for change detection in high resolution bi-temporal remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing 166 (2020), 183–200.Google ScholarCross Ref
- Hao Chen and Zhenwei Shi. 2020. A spatial-temporal attention-based method and a new dataset for remote sensing image change detection. Remote Sensing 12, 10 (2020), 1662.Google ScholarCross Ref
- Zhi Zheng, Yi Wan, Yongjun Zhang, Sizhe Xiang, Daifeng Peng, and Bin Zhang. 2021. 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
- Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, DavidWarde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2020. Generative adversarial networks. Commun. ACM 63, 11 (2020), 139–144.Google ScholarDigital Library
- Bo Fang, Gang Chen, Guichong Ouyang, Jifa Chen, Rong Kou, and Lizhe Wang. 2021. Content-Invariant Dual Learning for Change Detection in Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing (2021). DOI: 10.1109/TGRS.2021.3064501Google Scholar
- Hao Chen, Wenyuan Li, and Zhenwei Shi. 2021. Adversarial Instance Augmentation for Building Change Detection in Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing (2021). DOI: 10.1109/TGRS.2021.3066802Google Scholar
- Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, and Kaiming He. 2017. Aggregated residual transformations for deep neural networks. In Proceedings of the IEEE conference on computer vision and pattern recognition. 1492–1500.Google ScholarCross Ref
- Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár. 2017. Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision. 2980–2988.Google ScholarCross Ref
- Xiuwei Zhang, Yuanzeng Yue, Wenxiang Gao, Shuai Yun, Qian Su, Hanlin Yin, and Yanning Zhang. 2021. DifUnet++: A Satellite Images Change Detection Network Based on Unet++ and Differential Pyramid. IEEE Geoscience and Remote Sensing Letters (2021). DOI: 10.1109/LGRS.2021.3049370Google Scholar
- Zhaobin Cao, Mengmeng Wu, Rui Yan, Fa Zhang, and Xiaohua Wan. 2020. Detection of Small Changed Regions in Remote Sensing Imagery Using Convolutional Neural Network. In IOP Conference Series: Earth and Environmental Science, Vol. 502. IOP Publishing, 012017.DOI: 10.1088%2F1755-1315%2F502%2F1%2F012017Google Scholar
- Tao Lei, Yuxiao Zhang, Zhiyong Lv, Shuying Li, Shigang Liu, and Asoke K Nandi. 2019. Landslide inventory mapping from bitemporal images using deep convolutional neural networks. IEEE Geoscience and Remote Sensing Letters 16, 6 (2019), 982–986.40Google ScholarCross Ref
Recommendations
Multi-scale Dense Object Detection in Remote Sensing Imagery Based on Keypoints
Pattern Recognition and Computer VisionAbstractObject detection is a popular topic of computer vision and has attracted increasing attention in the field of remote sensing. However, object detection in remote sensing imagery is a very challenging task. Many existing detection methods depend on ...
Multi-scale ship target detection in remote sensing images
RICAI '20: Proceedings of the 2020 2nd International Conference on Robotics, Intelligent Control and Artificial IntelligenceIn this paper, a new method is proposed for multi-scale ship target detection and recognition in remote sensing image, which uses the change receptive field design of neural network to realize multi-scale target detection. The network structure is ...
Semi-supervised label propagation for multi-source remote sensing image change detection
AbstractRemote sensing image change detection remains a challenging task. Most existing approaches are based on fully supervised learning, but labeled data are so scarce for change detection. It is difficult to exhibit high detection ...
Highlights- Propose a clustering label propagation method to expand the labeled data.
- ...
Comments