Abstract
The tilted viewing nature of the off-nadir aerial images brings severe challenges to the building change detection (BCD) problem: the mismatch of the nearby buildings and the semantic ambiguity of the building facades. To tackle these challenges, we present a multi-task guided change detection network model, named as MTGCD-Net. The proposed model approaches the specific BCD problem by designing three auxiliary tasks, including: (1) a pixel-wise classification task to predict the roofs and facades of buildings; (2) an auxiliary task for learning the roof-to-footprint offsets of each building to account for the misalignment between building roof instances; and (3) an auxiliary task for learning the identical roof matching flow between bi-temporal aerial images to tackle the building roof mismatch problem. These auxiliary tasks provide indispensable and complementary building parsing and matching information. The predictions of the auxiliary tasks are finally fused to the main BCD branch with a multi-modal distillation module. To train and test models for the BCD problem with off-nadir aerial images, we create a new benchmark dataset, named BANDON. Extensive experiments demonstrate that our model achieves superior performance over the previous state-of-the-art competitors.
Similar content being viewed by others
References
Fujita A, Sakurada K, Imaizumi T, et al. Damage detection from aerial images via convolutional neural networks. In: Proceedings of the 15th IAPR International Conference on Machine Vision Applications (MVA), 2017. 5–8
Gupta R, Goodman B, Patel N, et al. Creating xBD: a dataset for assessing building damage from satellite imagery. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019. 10–17
Dueker K J, Horton F E. Urban-change detection systems: remote-sensing inputs. Photogrammetria, 1972, 28: 89–106
Chen J, Liu H, Hou J, et al. Improving building change detection in VHR remote sensing imagery by combining coarse location and co-segmentation. ISPRS Int J Geo-Inform, 2018, 7: 213
Awrangjeb M. Effective generation and update of a building map database through automatic building change detection from LiDAR point cloud data. Remote Sens, 2015, 7: 14119–14150
Nie G, Liao G, Zeng C. SAR image change detection method based on PPNN. Sci China Inf Sci, 2021, 64: 189304
Xu B W, Xu J K, Xue N, et al. Accurate polygonal mapping of buildings in satellite imagery. 2022. ArXiv:2208.00609
Ding J, Xue N, Long Y, et al. Learning RoI transformer for oriented object detection in aerial images. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019
Ding M T, Jin Z, Tian Z, et al. Object registration for remote sensing images using robust kernel pattern vectors. Sci China Inf Sci, 2012, 55: 2611–2623
Kang Y, Ding Y, Li Z R, et al. A networked remote sensing system for on-road vehicle emission monitoring. Sci China Inf Sci, 2017, 60: 043201
Liu F, Zhu J H, Wang W F, et al. Surface-to-air missile sites detection agent with remote sensing images. Sci China Inf Sci, 2021, 64: 194201
Xu F, Hu C, Li J, et al. Special focus on deep learning in remote sensing image processing. Sci China Inf Sci, 2020, 63: 140300
Daudt R C, Le Saux B, Boulch A. Fully convolutional siamese networks for change detection. In: Proceedings of the 25th IEEE International Conference on Image Processing (ICIP), 2018. 4063–4067
Chen H, Qi Z, Shi Z. Remote sensing image change detection with transformers. IEEE Trans Geosci Remote Sens, 2022, 60: 1–14
Chen H, Shi Z. A spatial-temporal attention-based method and a new dataset for remote sensing image change detection. Remote Sens, 2020, 12: 1662
Ji S, Wei S, Lu M. Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set. IEEE Trans Geosci Remote Sens, 2018, 57: 574–586
Sun Y, Zhang X, Huang J, et al. Fine-grained building change detection from very high-spatial-resolution remote sensing images based on deep multitask learning. IEEE Geosci Remote Sens Lett, 2022, 19: 1–5
Li W J, Meng L X, Wang J W, et al. 3D building reconstruction from monocular remote sensing images. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021. 12548–12557
Wang J, Meng L, Li W, et al. Learning to extract building footprints from off-nadir aerial images. IEEE Trans Pattern Anal Mach Intell, 2023, 45: 1294–1301
van Etten A, Hogan D, Manso J, et al. The multi-temporal urban development spacenet dataset. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021. 6398–6407
Yang K, Xia G S, Liu Z, et al. Asymmetric siamese networks for semantic change detection in aerial images. IEEE Trans Geosci Remote Sens, 2022, 60: 1–18
Shen L, Lu Y, Chen H, et al. S2Looking: a satellite side-looking dataset for building change detection. Remote Sens, 2021, 13: 5094
Dellinger F, Delon J, Gousseau Y, et al. Change detection for high resolution satellite images, based on SIFT descriptors and an a contrario approach. In: Proceedings of IEEE Geoscience and Remote Sensing Symposium, 2014. 1281–1284
Padron-Hidalgo J A, Laparra V, Longbotham N, et al. Kernel anomalous change detection for remote sensing imagery. IEEE Trans Geosci Remote Sens, 2019, 57: 7743–7755
Bu S, Li Q, Han P, et al. Mask-CDNet: a mask based pixel change detection network. Neurocomputing, 2020, 378: 166–178
Chen B, Chen Z, Deng L, et al. Building change detection with RGB-D map generated from UAV images. Neurocomputing, 2016, 208: 350–364
Bertinetto L, Valmadre J, Henriques J F, et al. Fully-convolutional siamese networks for object tracking. In: Proceedings of European Conference on Computer Vision, 2016. 850–865
Zhang C, Yue P, Tapete D, et al. A deeply supervised image fusion network for change detection in high resolution bi-temporal remote sensing images. ISPRS J Photogramm Remote Sens, 2020, 166: 183–200
Liu Y, Pang C, Zhan Z, et al. Building change detection for remote sensing images using a dual-task constrained deep siamese convolutional network model. IEEE Geosci Remote Sens Lett, 2021, 18: 811–815
Daudt R C, Le Saux B, Boulch A, et al. Multitask learning for large-scale semantic change detection. Comput Vision Image Underst, 2019, 187: 102783
Esfandiari M, Abdi G, Jabari S, et al. Building change detection in off-nadir images using deep learning. In: Proceedings of IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2021. 1347–1350
He K M, Gkioxari G, Dollár P, et al. Mask R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, 2017. 2961–2969
Zheng Z, Ma A L, Zhang L P, et al. Change is everywhere: single-temporal supervised object change detection in remote sensing imagery. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021. 15193–15202
Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015. 3431–3440
Sudre C H, Li W Q, Vercauteren T, et al. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Proceedings of Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 2017. 240–248
Zhao H S, Shi J P, Qi X J, et al. Pyramid scene parsing network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017. 2881–2890
He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016. 770–778
Bandara W G C, Patel V M. A transformer-based siamese network for change detection. In: Proceedings of IEEE International Geoscience and Remote Sensing Symposium IGARSS, 2022
Huang X, Serge B. Arbitrary style transfer in real-time with adaptive instance normalization. In: Proceedings of the IEEE International Conference on Computer Vision, 2017. 1501–1510
Ghiasi G, Zoph B, Cubuk E, et al. Multi-task self-training for learning general representations. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021. 8856–8865
Acknowledgements
This work was supported in part by National Nature Science Foundation of China (Grant Nos. 41820104006, U22B2011, 61922065) and National Key R&D Program of China (Grant No. 2021YFB3900503).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pang, C., Wu, J., Ding, J. et al. Detecting building changes with off-nadir aerial images. Sci. China Inf. Sci. 66, 140306 (2023). https://doi.org/10.1007/s11432-022-3691-4
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11432-022-3691-4