Page 174 - 《武汉大学学报(信息科学版)》2025年第9期
P. 174

1902                            武 汉 大 学 学 报  (信 息 科 学 版)                        2025 年 9 月

                [13]  ZHENG Z, ZHONG Y F, WANG J J, et al.  Build⁃   Pattern Analysis and Machine Intelligence, 2018, 40
                     ing  Damage  Assessment  for  Rapid  Disaster  Re⁃  (4): 834-848.
                     sponse with a Deep Object-Based Semantic Change   [24]  LUO  W  J,  LI  Y  J,  URTASUN  R,  et  al.   Under⁃
                     Detection  Framework:  From  Natural  Disasters  to   standing the Effective Receptive Field in Deep Con⁃
                     Man-Made  Disasters[J].   Remote  Sensing  of  Envi⁃  volutional  Neural  Networks[EB/OL].  [2024-10-
                     ronment, 2021, 265: 112636.                     16] https://arxiv. org/abs/1701. 04128v2.
                [14]  DURNOV  V.   Xview2  First  Place  Solution.  [EB/  [25]  KAUR  N,  LEE  C  C,  MOSTAFAVI  A,  et  al.
                     OL].  [2024-10-16]  https://github. com/DIUx-   Large-Scale  Building  Damage  Assessment  Using  a
                     xView/xView2_first_place.                       Novel Hierarchical Transformer Architecture on Satel⁃
                [15]  WEBER  E,  KANÉ  H.   Building  Disaster  Damage   lite  Images[J].   Computer ⁃ Aided  Civil  and  Infra⁃
                     Assessment in Satellite Imagery with Multi-Tempo⁃  structure Engineering, 2023, 38(15): 2072-2091.
                     ral  Fusion[EB/OL].   2020:  2004. 05525.   https://  [26]  CHEN  H,  NEMNI  E,  VALLECORSA  S,  et  al.
                     arxiv. org/abs/2004. 05525v1.                   Dual-Tasks  Siamese  Transformer  Framework  for
                [16]  SHEN  Y,  ZHU  S  J,  YANG  T,  et  al.   BDANet:   Building  Damage  Assessment[C]//IEEE  Interna⁃
                     Multiscale  Convolutional  Neural  Network  with   tional Geoscience and Remote Sensing Symposium,
                     Cross-Directional  Attention  for  Building  Damage   Kuala Lumpur, Malaysia, 2022.
                     Assessment from Satellite Images[J].  IEEE Tran ⁃  [27]  LIU  Z ,  LIN  Y  T ,  CAO  Y ,  et  al.   Swin  Trans⁃

                     sactions  on  Geoscience  and  Remote  Sensing,  2021,   former:  Hierarchical  Vision  Transformer  Using
                     60: 5402114.                                    Shifted  Windows [C]//IEEE/CVF  International
                [17]  DENG  L  W,  WANG  Y.   Post-Disaster  Building   Conference on Computer Vision (ICCV), Montreal,
                     Damage Assessment Based on Improved U-Net[J].    Canada, 2021.
                     Scientific Reports, 2022, 12(1): 15862.    [28]  GU  A,  DAO  T.   Mamba:  Linear-Time  Sequence
                [18]  WANG  J  Q,  GUO  H  N,  SU  X,  et  al.   PCDAS⁃  Modeling  with  Selective  State  Spaces[EB/OL].
                     Net: Position-Constrained Differential Attention Sia⁃  2023:  2312. 00752.   https://arxiv. org/abs/2312.
                     mese  Network  for  Building  Damage  Assessment  00752v2.
                    [J].   IEEE  Transactions  on  Geoscience  and  Remote   [29]  LIU Y, TIAN Y J, ZHAO Y Z, et al.  VMamba:
                     Sensing, 2024, 62: 5622318.                     Visual State Space Model[EB/OL].  [2024-10-16]
                [19]  Ku J, SEO, J, JEON, T.  Dual-HRNet for Building   https://arxiv. org/abs/2401. 10166v4.
                     Localization and Damage Classification.  [EB/OL].    [30]  WAN Z F, ZHANG P P, WANG Y H, et al.  Sig⁃
                    [2024-10-16] https://github. com/SIAnalytics/du⁃  ma:  Siamese  Mamba  Network  for  Multi-modal  Se⁃
                     al-hrnet/blob/master/figures/xView2_White_Paper   mantic  Segmentation [EB/OL].   [2024-10-16]
                     _SI_Analytics. pdf.                             https://arxiv. org/abs/2404. 04256v2.
                [20]  XIE H F, HU X Y, JIANG H W, et al.  BSSNet:   [31]  MA X P, ZHANG X K, PUN M O.  RS3Mamba:
                     Building Subclass Segmentation from Satellite Images   Visual State Space Model for Remote Sensing Image
                     Using  Boundary  Guidance  and  Contrastive  Learning  Semantic  Segmentation[J].   IEEE  Geoscience  and
                    [J].   IEEE  Journal  of  Selected  Topics  in  Applied   Remote Sensing Letters, 2024, 21: 6011405.
                     Earth Observations and Remote Sensing, 2022, 15:   [32]  BERMAN  M,  TRIKI  A  R,  BLASCHKO  M  B.
                     7700-7711.                                      The  Lovasz-Softmax  Loss:  A  Tractable  Surrogate
                [21]  GUPTA  R,  SHAH  M.   RescueNet:  Joint  Building   for  the  Optimization  of  the  Intersection-over-Union
                     Segmentation  and  Damage  Assessment  from  Satel⁃  Measure in Neural Networks[C]//IEEE/CVF Con⁃
                     lite  Imagery[C]//25th  International  Conference  on   ference  on  Computer  Vision  and  Pattern  Recogni⁃
                     Pattern Recognition (ICPR), Milan, Italy, 2021.  tion, Salt Lake City, USA, 2018.
                [22]  HE K M, ZHANG X Y, REN S Q, et al.  Deep Re⁃  [33]  Lee  C  C,  Kaur  N,  Mahdavi-Amiri  A,  et  al.   Ida-
                     sidual  Learning  for  Image  Recognition[C]//IEEE   BD: Pre- and Post-Disaster High-Resolution Satel⁃
                     Conference on Computer Vision and Pattern Recog⁃  lite  Imagery  for  Building  Damage  Assessment  from
                     nition (CVPR), Las Vegas, NV, USA, 2016.        Hurricane  Ida.   Designsafe-CI.  [EB/OL].  [2024-
                [23]  CHEN  L  C,  PAPANDREOU  G,  KOKKINOS  I,      10-16]  https://www.  designsafe-ci.  org/data/
                     et al.  DeepLab: Semantic Image Segmentation with   browser/public/designsafe. storage. published/ PRJ-
                     Deep Convolutional Nets, Atrous Convolution, and   3563.
                     Fully  Connected  CRFS[J].   IEEE  Transactions  on   [34]  CHEN  H,  SHI  Z  W.   A  Spatial-Temporal  Atten⁃
   169   170   171   172   173   174   175   176   177   178   179