предыдущий|следующий

Domain-adaptation algorithm for remotely sensing building changes through instance contrast learning

  • role： First author第一作者
  • Affiliation：

    National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi'an Jiaotong University, Xi'an 710049, China

    National Engineering Research Center for Visual Information and Applications, Xi'an Jiaotong University, Xi'an 710049, China

  • Email：a1182693164@stu.xjtu.edu.cn
  • Introduction：张奇，研究方向为遥感图像变化检测。E-mail：a1182693164@stu.xjtu.edu.cn
ZHANG Qi12,  
  • Affiliation：

    Beijing Institute of Remote Sensing, Beijing 100011, China

LU Yao3,  
  • Affiliation：

    National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi'an Jiaotong University, Xi'an 710049, China

    National Engineering Research Center for Visual Information and Applications, Xi'an Jiaotong University, Xi'an 710049, China

    Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an 710049, China

WANG Fei124,  
  • role： Corresponding author通信作者
  • Affiliation：

    National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi'an Jiaotong University, Xi'an 710049, China

    National Engineering Research Center for Visual Information and Applications, Xi'an Jiaotong University, Xi'an 710049, China

    Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an 710049, China

  • Email：xuetaozh@xjtu.edu.cn
  • Introduction：张雪涛，研究方向为模式识别与智能系统。E-mail： xuetaozh@xjtu.edu.cn
ZHANG Xuetao124*,  
  • Affiliation：

    National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Xi'an Jiaotong University, Xi'an 710049, China

    National Engineering Research Center for Visual Information and Applications, Xi'an Jiaotong University, Xi'an 710049, China

    Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, Xi'an 710049, China

ZHENG Nanning124

реферат

Building-change detection automatically identifies changes in ground buildings in remote-sensing images acquired in the same geographic area at different times. Fully supervised change-detection algorithms require a large amount of labeled remote sensing data to make accurate predictions. Manually labeling a building change detection label is time-consuming and labor intensive because it requires a professional to compare and label two images pixel by pixel. Unsupervised domain adaptation technique is an effective means to alleviate this problem. Although the current-domain adaptation algorithm has achieved good results in building-change detection, the following problems persist: Problem 1: A class-based domain mixing strategy is applicable to a large number of categories. In building-change detection, only positive samples of the category “change” are available. Problem 2: In the current pixel-based contrast learning method, pseudo labels generated by a model must have samples with classification errors because the labels of target domains are unidentifiable. This requirement introduces large noise information during contrast training. Problem 3: The pseudo label generated by high-confidence threshold filtering does not leverage the low confidence prediction results of a teacher model. To solve the above problems, this paper proposes a case-level contrast-learning domain-adaptation algorithm for cross-domain building-change detection task.This paper proposes an instance contrast-learning domain adaptation for change detection (ICDA-CD) method for cross-domain building-change detection. The main contributions are as follows: (1) A region-level domain-mixing method is proposed, which combines data containing the buildings in a source domain and data containing buildings in a target domain on one sample simultaneously. (2) Case-level contrast learning method is proposed. In the encoder, the distance between the biphasic features of a changing building area is pulled apart. In the decoder, the distance between the features of each changing building area is narrowed. (3) A pseudo label quality estimation method is proposed. The pseudo-label quality of each pixel position is estimated by the value predicted by a teacher model, and then loss is weighted.Domain migration experiments were performed on the LEVIR-CD and S2Looking datasets, and comparison and ablation experiments were performed with advanced domain-adaptation algorithms. In the migration of the LEVIR-CD task to the S2Looking task, the proposed algorithm achieved the highest F1 and IOU of 43.91 and 28.31, respectively. In the migration of the S2Looking task to the LEVIR-CD task, the proposed algorithm achieved the highest F1 scores and IOU of 74.75 and 59.68, respectively.To solve the problem of unsupervised domain adaptive change detection algorithm across data domains, an ICDA-CD method was proposed. The accuracy of the cross-domain unsupervised domain adaptive change detection algorithm was effectively improved by using region-level domain mixing, case-level contrast learning, and pseudo label quality estimation-weighted loss.

ключеви́че слова́

remote sensing images;building change detection;contrast learning;domain adaptation;deep learning;pseudo label

References

  1. 1.
    Arnaudo E, Tavera A, Masone C, Dominici F and Caputo B. 2023. Hierarchical instance mixing across domains in aerial segmentation. IEEE Access, 11: 13324-13333
  2. 2.
    Asokan A and Anitha J. 2019. Change detection techniques for remote sensing applications: a survey. Earth Science Informatics, 12(2): 143-160
  3. 3.
    Bandara W G C and Patel V M. 2022. A transformer-based Siamese network for change detection//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium. Kuala Lumpur: IEEE: 207-210
  4. 4.
    Chen H, Qi Z P and Shi Z W. 2021. Remote sensing image change detection with transformers. IEEE Transactions on Geoscience and Remote Sensing, 60: 5607514
  5. 5.
    Chen H and Shi Z W. 2020. A spatial-temporal attention-based method and a new dataset for remote sensing image change detection. Remote Sensing, 12(10): 1662
  6. 6.
    Chen H R X, Wu C, Du B, Zhang L P and Wang L. 2020. Change detection in multisource VHR images via deep Siamese convolutional multiple-layers recurrent neural network. IEEE Transactions on Geoscience and Remote Sensing, 58(4): 2848-2864
  7. 7.
    Chen H Y. 2022. Change Detection Technology of Heterogeneous Remote Sensing Images Based on Style Transfer. Guangzhou: Guangdong University of Technology
  8. 8.
    Chen M, Zheng Z D, Yang Y and Chua T S. 2023. PiPa: pixel-and patch-wise self-supervised learning for domain adaptative semantic segmentation//Proceedings of the 31st ACM International Conference on Multimedia. Ottawa: ACM: 1905-1914
  9. 9.
    Chen P, Zhang B, Hong D F, Chen Z C, Yang X and Li B P. 2022. FCCDN: Feature constraint network for VHR image change detection. ISPRS Journal of Photogrammetry and Remote Sensing, 187: 101-119
  10. 10.
    Cheng X and Lei H. 2023. Semantic segmentation of remote sensing imagery based on multiscale deformable CNN and DenseCRF. Remote Sensing, 15(5): 1229
  11. 11.
    Daudt R C, Le Saux B and Boulch A. 2018. Fully convolutional siamese networks for change detection//25th IEEE International Conference on Image Processing (ICIP). Athens: IEEE: 4063-4067
  12. 12.
    Fang S, Li K Y, Shao J Y and Li Z. 2022. SNUNet-CD: A densely connected Siamese network for change detection of VHR images. IEEE Geoscience and Remote Sensing Letters, 19: 8007805
  13. 13.
    Ganin Y and Lempitsky V. 2015. Unsupervised domain adaptation by backpropagation//Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille: JMLR.org: 1180-1189
  14. 14.
    Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A and Bengio Y. 2020. Generative adversarial networks. Communications of the ACM, 63(11): 139-144
  15. 15.
    Hoyer L, Dai D X and Van Gool L. 2022a. DAFormer: improving network architectures and training strategies for domain-adaptive semantic segmentation//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE: 9914-9925
  16. 16.
    Hoyer L, Dai D X and Van Gool L. 2022b. HRDA: context-aware high-resolution domain-adaptive semantic segmentation//Proceedings of the 17th European Conference on Computer Vision. Tel Aviv: Springer: 372-391
  17. 17.
    Hoyer L, Dai D X, Wang H R and Van Gool L. 2023. MIC: masked image consistency for context-enhanced domain adaptation//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver: IEEE: 11721-11732
  18. 18.
    Huang S T, Wang D L, Wu X H and Tang A. 2019. DSANet: Dual self-attention network for multivariate time series forecasting//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Beijing: ACM: 2129-2132
  19. 19.
    Khankeshizadeh E, Mohammadzadeh A, Moghimi A and Mohsenifar A. 2022. FCD-R2U-net: Forest change detection in bi-temporal satellite images using the recurrent residual-based U-net. Earth Science Informatics, 15(4): 2335-2347
  20. 20.
    Li G R, Kang G L, Liu W, Wei Y C and Yang Y. 2020. Content-consistent matching for domain adaptive semantic segmentation//Proceedings of the 16th European Conference on Computer Vision. Glasgow: Springer: 440-456
  21. 21.
    Liu H C and Zhang L. 2020. Adaptive threshold change detection based on type feature for remote sensing image. Journal of Remote Sensing (in Chinese), 24(6): 728-738
  22. 22.
    Liu S C, Du K C, Zheng Y J, Chen J, Du P J and Tong X H. 2023. Remote sensing change detection technology in the era of artificial intelligence: inheritance, development and challenges. Journal of Remote Sensing (in Chinese), 27(9): 1975-1987
  23. 23.
    Liu X F, Yoo C, Xing F X, Oh H, El Fakhri G, Kang J W and Woo J. 2022a. Deep unsupervised domain adaptation: A review of recent advances and perspectives. APSIPA Transactions on Signal and Information Processing, 11(1): e25
  24. 24.
    Liu X G, Li M M, Wang X Q and Zhang Z C. 2024. Use of object-based Siamese neural network to build change detection from very high resolution remote-sensing images. National Remote Sensing Bulletin, 28(2): 437-454
  25. 25.
    Liu Y, He X, Li D Y, Yue H and Wei J L. 2024. CBAM UNet+++: attention mechanism to guide change detection study of full-scale connected networks. National Remote Sensing Bulletin, 28(4)： 1052-1065
  26. 26.
    Liu Z, Mao H Z, Wu C Y, Feichtenhofer C, Darrell T and Xie S N. 2022b. A convnet for the 2020s//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE: 11966-11976
  27. 27.
    Olsson V, Tranheden W, Pinto J and Svensson L. 2021. ClassMix: segmentation-based data augmentation for semi-supervised learning//Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision. Waikoloa: IEEE: 1368-1377
  28. 28.
    Shen L, Lu Y, Chen H, Wei H, Xie D H, Yue J B, Chen R, Lv S Y and Jiang B. 2021. S2Looking: a satellite side-looking dataset for building change detection. Remote Sensing, 13(24): 5094
  29. 29.
    Shi W X, Tan D L and Bao S L. 2020. Feature enhancement SSD algorithm and its application in remote sensing images target detection. Acta Photonica Sinica, 49(1): 0128002
  30. 30.
    Shi W Z, Zhang M, Zhang R, Chen S X and Zhan Z. 2020. Change detection based on artificial intelligence: state-of-the-art and challenges. Remote Sensing, 12(10): 1688
  31. 31.
    Su Y Z, Cheng J, Bai H W, Liu H J and He C T. 2022. Semantic segmentation of very-high-resolution remote sensing images via deep multi-feature learning. Remote Sensing, 14(3): 533
  32. 32.
    Tranheden W, Olsson V, Pinto J and Svensson L. 2021. DACS: Domain adaptation via cross-domain mixed sampling//Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision. Waikoloa: IEEE: 1378-1388
  33. 33.
    Tsai Y H, Hung W C, Schulter S, Sohn K, Yang M H and Chandraker M. 2018. Learning to adapt structured output space for semantic segmentation//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE: 7472-7481
  34. 34.
    Vayyat M, Kasi J, Bhattacharya A, Ahmed S and Tallamraju R. 2022. CLUDA: contrastive learning in unsupervised domain adaptation for semantic segmentation. arXiv preprint arXiv:2208.14227
  35. 35.
    Vu T H, Jain H, Bucher M, Cord M and Pérez P. 2019. ADVENT: adversarial entropy minimization for domain adaptation in semantic segmentation//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE: 2512-2521
  36. 36.
    Wang Y X, Peng J R and Zhang Z X. 2021. Uncertainty-aware pseudo label refinery for domain adaptive semantic segmentation//2021 IEEE/CVF International Conference on Computer Vision. Montreal: IEEE: 9072-9081
  37. 37.
    Xiao T T, Liu Y C, Zhou B L, Jiang Y N and Sun J. 2018. Unified perceptual parsing for scene understanding//Proceedings of the 15th European Conference on Computer Vision. Munich: Springer: 432-448
  38. 38.
    Xue C C, Gao C X, Hu J, Qiu S and Wang Q. 2023. Marine boat lights detection method based on sparse representation for low light remote sensing images during night. Journal of Electronics & Information Technology, 45(5): 1747-1757
  39. 39.
    Yang Y C and Soatto S. 2020. FDA: Fourier domain adaptation for semantic segmentation//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE: 4084-4094
  40. 40.
    Yao Q L, Hu X and Lei H. 2019. Object detection in remote sensing images using multiscale convolutional neural networks. Acta Optica Sinica, 39(11): 1128002
  41. 41.
    Zhou C J, Wang H F, Chen X F and Yang Y. 2024. Detection of earthquake-damaged buildings via UAV high-resolution remote sensing images. National Remote Sensing Bulletin, 28(4)： 911-925

Читать полностью

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website