随着遥感技术和城市化进程的快速发展，基于遥感影像的建筑物变化检测在城市规 划、灾害监测、资源管理等领域中发挥着关键作用。然而在实际应用中，由于遥感影像 成像过程中受传感器类型、成像时间差、气象条件、观测角度变化及地物复杂性等多种 因素影响，传统变化检测方法往往难以保证检测精度与鲁棒性，而深度学习技术的引入 为变化检测任务提供了更强的特征提取与表达能力。本文围绕遥感影像建筑物变化检测 的关键难点，采用深度学习技术，通过构建相关的方法模型，进一步提升检测质量。论 文的主要研究内容如下： （1）针对建筑物变化检测中因季节变化、光照差异等因素所导致的边界模糊与小型 建筑漏检问题，本文提出了一种融合多尺度混合注意力机制的变化检测网络模型。该模 型设计了双分支特征融合模块，以充分挖掘双时相遥感图像中的变化信息。同时引入了 由通道和空间注意力机制组成的混合注意力模块，有效提升了模型对关键变化区域的关 注能力。利用跳跃连接结构融合深层语义特征与浅层细节特征，增强了模型的判别能力 与变化区域的边界清晰度。模型在LEVIR-CD与WHU-CD两个公开数据集分别获得了 91.58%和 91.81%的 F1 分数，结果优于多种现有主流方法，验证了其在变化检测任务中 的优越性能。 （2）针对建筑物变化检测中因双时相影像成像视角不同所引发的伪变化问题，本文 提出了一种基于多任务的语义特征约束变化检测网络模型。该模型采用一种交换双编码 器-解码器结构作为主干网络，通过融合双时相的语义与空间特征信息，提升了模型对实 际变化的判别能力。针对因成像视角差异所引起的双时相特征空间错位问题，引入了基 于密集连接的特征融合模块，从结构上增强了模型对视角差异的建模能力，有效提升了 特征对齐精度。同时设计了轻量级增强卷积模块，用来进一步优化特征提取效果。模型 不仅在LEVIR-CD 与WHU-CD两个数据集保持了良好的基础性能，而且在成像视角差 异明显的公开数据集NJDS和自制数据集ZK-CD上分别获得了74.35%和70.93%的F1 分数，显示出其在处理由成像视角差异导致的伪变化问题时所具有的显著优势。 （3）为满足用户在不同应用场景下对建筑物变化检测方法的多样化需求，并提升检 测流程的智能化和效率，本文构建了一个基于PyQt5的建筑物变化检测系统，系统集成 了以上两种面向不同场景的深度学习变化检测网络模型，具备完整的数据处理与检测功 能。 综上所述，本文针对建筑物变化检测任务中存在的多种复杂场景，提出了两种深度 学习模型以提升变化检测的精度，同时设计并实现了一个变化检测系统，有效增强了检 测的便捷性与实用性。

With the rapid development of remote sensing technology and urbanization, building change detection based on remote sensing images plays a key role in urban planning, disaster monitoring, resource management and other fields. However, in practical applications, due to the imaging process of remote sensing images is affected by many factors such as sensor type, imaging time difference, meteorological conditions, observation angle changes and complexity of features, traditional change detection methods are often difficult to ensure the detection accuracy and robustness, and the introduction of deep learning technology provides a stronger feature extraction and expression capability for the change detection task. This paper focuses on the key difficulties of building change detection in remote sensing images, adopts deep learning technology, and further improves the detection quality by constructing relevant method models. The main research content of the paper is as follows: (1) Aiming at the problems of boundary blurring and small building leakage due to seasonal changes and lighting differences in building change detection, this paper proposes a change detection network model that incorporates a multi-scale hybrid attention mechanism. The model is designed with a dual-branch feature fusion module to fully mine the change information in dual-temporal remote sensing images. A hybrid attention module consisting of channel and spatial attention mechanisms is also introduced to effectively enhance the model's ability to focus on key change regions. The skip connection structure is utilized to fuse the deep semantic features with the shallow detail features, which enhances the model's discriminative ability and the boundary clarity of the change regions. The model obtains F1 scores of 91.58% and 91.81% in two public datasets, LEVIR-CD and WHU-CD, respectively, and the results outperform many existing mainstream methods, verifying its superior performance in the change detection task. (2) Aiming at the pseudo-change problem in building change detection caused by the different imaging viewpoints of dual-time-phase images, this paper proposes a multi-task-based semantic feature-constrained change detection network model. The model adopts a switched dual encoder-decoder structure as the backbone network, which improves the model's ability to discriminate actual changes by fusing the semantic and spatial feature information of dual-temporal phase. For the problem of spatial misalignment of dual-temporal-phase features caused by differences in imaging viewpoints, a feature fusion module based on dense connectivity is introduced, which structurally enhances the model's ability of modeling differences in viewpoints, and effectively improves the feature alignment accuracy. A lightweight enhanced convolution module is also designed and used to further optimize the feature extraction effect. The model not only maintains good basic performance on the LEVIR-CD and WHU-CD datasets, but also obtains F1 scores of 74.35% and 70.93% on the publicly available dataset NJDS and the homemade dataset ZK-CD, which have obvious differences in imaging viewpoints, respectively, which shows its significant advantages in dealing with pseudo-variation problems caused by differences in imaging viewpoints. (3) In order to meet the diversified needs of users for building change detection methods in different application scenarios and to enhance the intelligence and efficiency of the detection process, this paper constructs a building change detection system based on PyQt5, which integrates the above two deep learning change detection network models for different scenarios and has complete data processing and detection functions. In summary, this paper proposes two deep learning models to improve the accuracy of change detection for the multiple complex scenarios existing in the building change detection task, and at the same time, designs and implements a change detection system, which effectively enhances the convenience and practicality of the detection.

[1] Smith I A, Winbourne J B, Tieskens K F, et al. A satellite-based model for estimating latent heat flux from urban vegetation[J]. Frontiers in Ecology and Evolution, 2021, 9: 695995.

[2] 朱旭阳, 韦春桃, 何蔚, 等. 结合波段选择与改进 Shufflenetv2 的土地利用变化检测 [J]. 测绘地理信息, 2023, 48(06): 88-93.

[3] He H, Yan J, Liang D, et al. Time-series land cover change detection using deep learning-based temporal semantic segmentation[J]. Remote Sensing of Environment, 2024, 305: 114101.

[4] 李家艺, 黄昕, 胡宇平, 等.夜光影像和高分辨率影像耦合的土耳其Mw 7.8地震建筑倒塌智能解译[J]. 武汉大学学报(信息科学版), 2023, 48(10): 1706-1714.

[5] 冯永玖, 李鹏朔, 童小华, 等. 城市典型要素遥感智能监测与模拟推演关键技术[J]. 测绘学报, 2022, 51(04): 577-586.

[6] 张祖勋, 姜慧伟, 庞世燕, 等. 多时相遥感影像的变化检测研究现状与展望[J]. 测绘学报, 2022, 51(07): 1091-1107.

[7] Huang X, Zhang L, Zhu T. Building change detection from multitemporal high-resolution remotely sensed images based on a morphological building index[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013, 7(1): 105-115.

[8] Li X, Yeh A G O. Principal component analysis of stacked multi-temporal images for the monitoring of rapid urban expansion in the Pearl River Delta[J]. International Journal of Remote Sensing, 1998, 19(8): 1501-1518..

[9] Lu D, Mausel P, Brondizio E, et al. Change detection techniques[J]. International journal of remote sensing, 2004, 25(12): 2365-2401.

[10] 史芳行, 朱大明，程仕亿. 基于卷积神经网络的建筑物语义分割和变化检测研究[D]. 昆明理工大学, 2022.

[11] Bai B, Fu W, Lu T, et al. Edge-Guided Recurrent Convolutional Neural Network for Multitemporal Remote Sensing Image Building Change Detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 1-13.

[12] 周媛, 陈占龙，赵雷. 基于深度学习的高分辨率遥感影像建筑物提取与变化检测方法研究[D]. 中国地质大学, 2022.

[13] Khabou N, Rodriguez I B, Gharbi G, et al. A threshold based context change detection in pervasive environments: Application to a smart campus[J]. Procedia Computer Science, 2014, 32: 461-468.

[14] He P, Shi W, Zhang H, et al. A novel dynamic threshold method for unsupervised change detection from remotely sensed images[J]. Remote Sensing Letters, 2014, 5(4): 396-403.

[15] Lv Z Y, Shi W, Zhang X, et al. Landslide inventory mapping from bitemporal high- resolution remote sensing images using change detection and multiscale segmentation[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(5): 1520-1532.

[16] Yang G, Li H C, Wang W Y, et al. Unsupervised change detection based on a unified framework for weighted collaborative representation with RDDL and fuzzy clustering[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(11): 8890-8903.

[17] Gong M, Zhou Z, Ma J. Change detection in synthetic aperture radar images based on image fusion and fuzzy clustering[J]. IEEE Transactions on Image Processing, 2011, 21(4): 2141-2151.

[18] Hao M, Hua Z, Li Z, et al. Unsupervised change detection using a novel fuzzy c-means clustering simultaneously incorporating local and global information[J]. Multimedia Tools and Applications,2017, 76: 20081-20098.

[19] Lv Z, Liu T, Wan Y, et al. Post-processing approach for refining raw land cover change detection of very high-resolution remote sensing images[J]. Remote Sensing, 2018, 10(3): 472.

[20] Zanetti M, Bovolo F, Bruzzone L. Rayleigh-rice mixture parameter estimation via EM algorithm for change detection in multispectral images[J]. IEEE Transactions on Image Processing, 2015, 24(12): 5004-5016.

[21] Canty M J, Nielsen A A. Automatic radiometric normalization of multitemporal satellite imagery with the iteratively re-weighted MAD transformation[J]. Remote sensing of environment, 2008, 112(3): 1025-1036.

[22] Xu Q, Liu Z, Li F, et al. The regularized iteratively reweighted object-based MAD method for change detection in bi-temporal, multispectral data[C]//Hyperspectral Remote Sensing Applications and Environmental Monitoring and Safety Testing Technology. SPIE, 2016, 10156: 195-202.

[23] Fang S, Li K, Shao J, et al. SNUNet-CD: A Densely Connected Siamese Network for Change Detection of VHR Images[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 19: 1-5.

[24] Shao P, Shi W, He P, et al. Novel approach to unsupervised change detection based on a robust semi-supervised FCM clustering algorithm[J]. Remote Sensing, 2016, 8(3): 264.

[25] Sharma A, Gulati T. Change detection in remotely sensed images based on image fusion and fuzzy clustering[J]. International Journal of Electronics Engineering Research, 2017, 9(1): 141-150.

[26] Li H, Jing L, Wang L, et al. Semi-automatic mapping of large shallow landslides using high-resolution remote sensing imagery[C]//2016 4th International Workshop on Earth Observation and Remote Sensing Applications (EORSA). IEEE, 2016: 241-244.

[27] Lei T, Xue D, Lv Z, et al. Unsupervised change detection using fast fuzzy clustering for landslide mapping from very high-resolution images[J]. Remote Sensing, 2018, 10(9): 1381.

[28] Blaschke T, Kelly M, Merschdorf H. Object-based image analysis: Evolution, history,

state of the art, and future vision[J]. Remote Sensing Handbook, Volume II, 2024:

145-167.

[29] Zakeri F, Huang B, Saradjian M R. Fusion of change vector analysis in posterior

probability space and postclassification comparison for change detection from

multispectral remote sensing data[J]. Remote Sensing, 2019, 11(13): 1511.

[30] Yu H, Yang W, Hua G, et al. Change detection using high resolution remote sensing

images based on active learning and Markov random fields[J]. Remote Sensing, 2017,

9(12): 1233.

[31] Li Z, Shi W, Lu P, et al. Landslide mapping from aerial photographs using change

detection-based Markov random field[J]. Remote sensing of environment, 2016, 187:

76-90.

[32] Touati R, Mignotte M, Dahmane M. Multimodal change detection in remote sensing

images using an unsupervised pixel pairwise-based Markov random field model[J]. IEEE

Transactions on Image Processing, 2019, 29: 757-767.

[33] Yang Y, Cong X, Long K, et al. MRF model-based joint interrupted SAR imaging and

coherent change detection via variational Bayesian inference[J]. Signal Processing, 2018,

151: 144-154.

[34] Li Z, Shi W, Myint S W, et al. Semi-automated landslide inventory mapping from bitemporal aerial photographs using change detection and level set method[J]. Remote Sensing of Environment, 2016, 175: 215-230.

[35] Zhang X, Shi W, Hao M, et al. Level set incorporated with an improved MRF model for unsupervised change detection for satellite images[J]. European Journal of Remote Sensing, 2017, 50(1): 202-210.

[36] Zhang X, Xiao P, Feng X, et al. Separate segmentation of multi-temporal high-resolution remote sensing images for object-based change detection in urban area[J]. Remote

Sensing of Environment, 2017, 201: 243-255.

[37] 杨彬, 毛银, 陈晋, 等. 深度学习的遥感变化检测综述: 文献计量与分析[J]. 遥感学

报, 2023, 27(9): 1988-2005.

[38] 任秋如, 杨文忠, 汪传建. 遥感影像变化检测综述[J]. 计算机应用, 2021, 41(8):

2294-2305.

[39] Gong M, Yang H, Zhang P. Feature learning and change feature classification based on

deep learning for ternary change detection in SAR images[J]. ISPRS Journal of

Photogrammetry and Remote Sensing, 2017, 129: 212-225.

[40] Alcantarilla P F, Stent S, Ros G, et al. Street-view change detection with deconvolutional

networks[J]. Autonomous Robots, 2018, 42: 1301-1322.

[41] Daudt R C, Le Saux B, Boulch A. Fully convolutional siamese networks for change

detection[C]//2018 25th IEEE international conference on image processing (ICIP). IEEE,

2018: 4063-4067.

[42] Zheng Z, Wan Y, Zhang Y, et al. CLNet: Cross-layer convolutional neural network for CD

in optical remote sensing imagery[J]. ISPRS Journal of Photogrammetry and Remote

Sensing, 2021, 175:247-267.

[43] Varghese A, Gubbi J, Ramaswamy A, et al. ChangeNet: A deep learning architecture for

visual change detection[C]//Proceedings of the European conference on computer vision

(ECCV) workshops. 2018: 0-0.

[44] Han P, Ma C, Li Q, Leng P, Bu S, Li K. Aerial image CD using dual regions of interest

networks[J]. Neurocomputing, 2019, 349: 190-201.

[45] He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convolutional networks for

visual recognition[J]. IEEE transactions on pattern analysis and machine intelligence,

2015, 37(9): 1904-1916.

[46] Chen L C, Papandreou G, Kokkinos I, et al. Deeplab: Semantic image segmentation with

deep convolutional nets, atrous convolution, and fully connected crfs[J]. IEEE

transactions on pattern analysis and machine intelligence, 2017, 40(4): 834-848.

[47] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention

module[C]//Proceedings of the European conference on computer vision (ECCV). 2018:

3-19.

[48] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE

conference on computer vision and pattern recognition. 2018: 7132-7141.

[49] Wang Q, Wu B, Zhu P, et al. ECA-Net: Efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF conference on computer vision and

pattern recognition. 2020: 11534-11542.

[50] Chen H, Shi Z. A spatial-temporal attention-based method and a new dataset for remote

sensing image change detection[J]. Remote sensing, 2020, 12(10): 1662.

[51] Lyu H, Lu H, Mou L. Learning a transferable change rule from a recurrent neural network

for land cover change detection[J]. Remote Sensing, 2016, 8(6): 506.

[52] Chen Y, Ouyang X, Agam G. ChangeNet: Learning to detect changes in satellite

images[C]//Proceedings of the 3rd ACM SIGSPATIAL International Workshop on AI for

Geographic Knowledge Discovery. 2019: 24-31.

[53] Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[J].

Communications of the ACM, 2020, 63(11): 139-144.

[54] Ji S, Shen Y, Lu M, et al. Building instance change detection from large-scale aerial

images using convolutional neural networks and simulated samples[J]. Remote Sensing,

2019, 11(11): 1343.

[55] Wang Q, Zhang X, Chen G, et al. CD based on Faster R-CNN for high-resolution remote

sensing images[J]. Remote Sensing Letters, 2018, 9 (10-12): 923-932.

[56] Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic

segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern

recognition. 2015: 3431-3440.

[57] Lei T, Zhang Y, Lv Z, et al. Landslide inventory mapping from bitemporal images using

deep convolutional neural networks[J]. IEEE Geoscience and Remote Sensing Letters,

2019, 16(6): 982-986.

[58] Zhou Z, Siddiquee MMR, Tajbakhsh N, et al. UNet++: A nested u-net architecture for

medical image segmentation[M]//Deep learning in medical image analysis and

multimodal learning for clinical decision support. Springer, Cham, 2018: 3-11.

[59] Peng D, Zhang Y, Guan H. End-to-end change detection for high resolution satellite

images using improved UNet++[J]. Remote Sensing, 2019, 11(11): 1382.

[60] Bromley J, Guyon I, LeCun Y, et al. Signature verification using a" siamese" time delay

neural network[J]. Advances in Neural Information Processing Systems, 1993, 6:

737-744.

[61] 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[J]. ISPRS Journal of

Photogrammetry and Remote Sensing, 2020, 166: 183-200.

[62] Chen H, Wu C, Du B, et al. Change detection in multisource VHR images via deep Siamese convolutional multiple-layers recurrent neural network[J]. IEEE Transactions on

Geoscience and Remote Sensing, 2019, 58(4): 2848-2864.

[63] Duan Z, Wang S, Di H, et al. Distillation remote sensing object counting via multi-scale

context feature aggregation[J]. IEEE Transactions on Geoscience and Remote Sensing,

2021, 60: 1-12

[64] Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16x16 words:

Transformers for image recognition at scale[J]. arXiv preprint arXiv:2010.11929, 2020.

[65] Chen H, Qi Z, Shi Z. Remote sensing image change detection with transformers[J]. IEEE

Transactions on Geoscience and Remote Sensing, 2021, 60: 1-14.

[66] Wang Z, Zhang Y, Luo L, et al. TransCD: Scene change detection via transformer-based

architecture[J]. Optics Express, 2021, 29(25): 41409-41427.

[67] Zhang C, Wang L, Cheng S, et al. SwinSUNet: Pure transformer network for remote

sensing image change detection[J]. IEEE Transactions on Geoscience and Remote

Sensing, 2022, 60: 1-13.

[68] Bandara W G C, Patel V M. A transformer-based siamese network for change

detection[C]//IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing

Symposium. IEEE, 2022: 207-210.

[69] Gardner M W, Dorling S R. Artificial neural networks (the multilayer perceptron)—a

review of applications in the atmospheric sciences[J]. Atmospheric environment, 1998,

32(14-15): 2627-2636.

[70] Chen Y, Li W, Sakaridis C, et al. Domain adaptive faster r-cnn for object detection in the

wild[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.

2018: 3339-3348.

[71] 陈永. 基于深度学习的高分辨率遥感影像建筑物提取和变化检测研究[D]. 南京信息

工程大学, 2021.

[72] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//

Proceedings of the IEEE conference on computer vision and pattern recognition. 2016:

770-778.

[73] Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural

networks[C]//International conference on machine learning. PMLR, 2019: 6105-6114.

[74] 姜明, 张新长, 孙颖, 等.全尺度特征聚合的高分辨率遥感影像变化检测网络[J]. 测

绘学报, 2023, 52 (10): 1738-1748.

[75] Chen H, Song J, Han C, et al. Changemamba: Remote sensing change detection with

spatio-temporal state space model[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62:1-20.

[76] Liu Y, Shao Z, Hoffmann N. Global attention mechanism: Retain information to enhance

channel-spatial interactions[J]. arXiv preprint arXiv:2112.05561, 2021.

[77] He K, Zhang X, Ren S, et al. Delving deep into rectifiers: Surpassing human-level

performance on imagenet classification[C]//Proceedings of the IEEE international

conference on computer vision. 2015: 1026-1034.

[78] Wang Q, Wu B, Zhu P, et al. ECA-Net: Efficient channel attention for deep convolutional

neural networks[C]//Proceedings of the IEEE/CVF conference on computer vision and

pattern recognition. 2020: 11534-11542.

[79] De Boer P T, Kroese D P, Mannor S, et al. A tutorial on the cross-entropy method[J].

Annals of operations research, 2005, 134: 19-67.

[80] Ji S, Wei S, Lu M. Fully convolutional networks for multisource building extraction from

an open aerial and satellite imagery data set[J]. IEEE Transactions on geoscience and

remote sensing, 2018, 57(1): 574-586.

[81] Fang S, Li K, Shao J, et al. SNUNet-CD: A densely connected Siamese network for

change detection of VHR images[J]. IEEE Geoscience and Remote Sensing Letters, 2021,

19: 1-5.

[82] Zhang M, Shi W. A feature difference convolutional neural network-based change

detection method[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(10):

7232-7246.

[83] Ji S, Shen Y, Lu M, et al. Building instance change detection from large-scale aerial

images using convolutional neural networks and simulated samples[J]. Remote Sensing,

2019, 11(11): 1343.

[84] Daudt R C, Le Saux B, Boulch A, et al. Multitask learning for large-scale semantic

change detection[J]. Computer Vision and Image Understanding, 2019, 187: 102783.

[85] Papadomanolaki M, Karantzalos K, Vakalopoulou M. A multi-task deep learning

framework coupling semantic segmentation and image reconstruction for very high

resolution imagery[C]//IGARSS 2019-2019 IEEE International Geoscience and Remote

Sensing Symposium. IEEE, 2019: 1069-1072.

[86] Yang K, Xia G S, Liu Z, et al. Asymmetric siamese networks for semantic change

detection in aerial images[J]. IEEE Transactions on Geoscience and Remote Sensing,

2021, 60: 1-18.

[87] Tsutsui S, Hirakawa T, Yamashita T, et al. Semantic segmentation and change detection by

multi-task U-Net[C]//2021 IEEE International Conference on Image Processing (ICIP). IEEE, 2021: 619-623.

[88] 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[J]. ISPRS Journal of

Photogrammetry and Remote Sensing, 2020, 166: 183-200.

[89] Zhang L, Hu X, Zhang M, et al. Object-level change detection with a dual correlation

attention-guided detector[J]. ISPRS Journal of Photogrammetry and Remote Sensing,

2021, 177: 147-160.

[90] Chen P, Zhang B, Hong D, et al. FCCDN: Feature constraint network for VHR image

change detection[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 187:

101-119.

[91] Shen Q, Huang J, Wang M, et al. Semantic feature-constrained multitask siamese network

for building change detection in high-spatial-resolution remote sensing imagery[J]. ISPRS

Journal of Photogrammetry and Remote Sensing, 2022, 189: 78-94.

[92] Buslaev A, Iglovikov V I, Khvedchenya E, et al. Albumentations: fast and flexible image

augmentations[J]. Information, 2020, 11(2): 125.

[93] Kendall A, Gal Y, Cipolla R. Multi-task learning using uncertainty to weigh losses for

scene geometry and semantics[C]//Proceedings of the IEEE conference on computer

vision and pattern recognition. 2018: 7482-7491.

[94] Zhao H, Shi J, Qi X, et al. Pyramid scene parsing network[C]//Proceedings of the IEEE

conference on computer vision and pattern recognition. 2017: 2881-2890.

[95] 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[J]. IEEE Geoscience

and Remote Sensing Letters, 2020, 18(5): 811-815.