Please wait a minute...
 
Remote Sensing for Natural Resources    2025, Vol. 37 Issue (6) : 77-87     DOI: 10.6046/zrzyyg.2024242
|
DN-NET: A densely nested network for building extraction from remote sensing images
LIU Yi(), LIU Tao, GAO Tianying(), LI Guoyan
School of Computer and Information Engineering, Tianjin Chengjian University, Tianjin 300384, China
Download: PDF(6827 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks    
Abstract  

Building extraction aims to separate building pixels from remote sensing images, which plays a crucial role in applications such as urban planning and urban dynamic monitoring. However, building extraction generally faces challenges, such as void, false positives, and false negatives. Given this, this paper proposed a densely nested network (DN-Net). The sub-networks in the DN-Net were integrated with the enhanced residual convolutional module (ERCM) to extract rough contours of buildings from remote sensing images. Furthermore, to accurately locate the buildings, a coordinate attention module (CAM) was incorporated, effectively avoiding false positives. To deal with the holes during building extraction, a cascade convolutional module (CCM) was used, allowing the extraction of richer details with convolution kernels of various sizes, thereby ensuring accurate building extraction. The DN-Net was tested with the WHU datasets to assess its accuracy. The results showed that the DN-Net exhibited an intersection over union (IoU) of 89.20% and a F1 score of 94.29% on the validation set and 89.85% and 94.65%, respectively, on the test set. The results confirm that the DN-Net can significantly improve the building extraction accuracy, with more complete and detailed boundaries of buildings being extracted, demonstrating an outstanding ability to extract buildings of varying sizes.
Keywords building extraction      enhanced residual convolutional module (ERCM)      coordinate attention module (CAM)      cascade convolutional module (CCM)     
ZTFLH:  TP79  
Issue Date: 31 December 2025
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Yi LIU
Tao LIU
Tianying GAO
Guoyan LI
Cite this article:   
Yi LIU,Tao LIU,Tianying GAO, et al. DN-NET: A densely nested network for building extraction from remote sensing images[J]. Remote Sensing for Natural Resources, 2025, 37(6): 77-87.
URL:  
https://www.gtzyyg.com/EN/10.6046/zrzyyg.2024242     OR     https://www.gtzyyg.com/EN/Y2025/V37/I6/77
Fig.1  DN-Net network model
Fig.2  Processing process of sub module
Fig.3  Differential convolutional block
Fig.4  Coordinate attention module
Fig.5  Cascade convolutional module
训练优化法 IoU F1 P R
Adam 86.92 93.00 92.19 93.82
Adagrad 88.02 93.63 93.58 93.53
SGD 89.20 94.29 94.31 94.27
Tab.1  Results of different training optimization methods (%)
Fig.6  Schematic diagram of deep supervision
深度监督 IoU F1 P R
${X}_{\mathrm{D}\mathrm{e}}^{\mathrm{0,5}}$ 88.79 94.04 92.91 95.23
DS2 88.10 93.67 92.45 94.92
DS1 89.20 94.29 94.31 94.27
Tab.2  Impact of deep supervision on the segmentation performance of DN-Net (%)
参数 DN-Net Block3 CAM CCM IoU F1 P
表现 √ √ — — 88.82 94.08 93.46
√ √ √ — 89.02 94.19 93.99
√ √ — √ 89.09 94.23 94.11
√ √ √ √ 89.20 94.29 94.31
Tab.3  CAM and CCM feature fusion ablation experiments (%)
Fig.7  Visualization of feature response
类型 小型建筑物 密集建筑物 大型建筑物
示例1 示例2 示例1 示例2 示例1 示例2
原图
标签
DN-Net +Block3
DN-Net +Block3+CAM
DN-Net +Block3+CCM
DN-Net+Block3+CAM+CCM
Tab.4  Visual comparison
卷积块 IoU F1 P R
Block1 88.69 94.00 93.79 94.22
Block2 88.78 94.10 93.72 94.60
Block3 88.82 94.08 93.46 94.70
Tab.5  Improving residual convolution results (%)
卷积核 IoU F1 P R
[1,3] 88.68 94.00 93.34 94.68
[1,3,5] 88.82 94.08 93.46 94.70
[1,3,5,7] 89.09 94.23 94.11 94.35
Tab.6  CCM module ablation experiment (%)
网络 验证集 测试集
IoU F1 P R IoU F1 P R
SegNet 87.39 92.27 93.15 93.39 88.07 93.66 93.69 93.62
UNet 87.22 93.17 92.48 93.88 88.24 93.75 94.43 93.09
ENet 85.79 92.35 91.62 93.09 86.62 92.83 93.51 92.17
UNet++ 88.18 93.72 93.51 93.92 88.65 93.99 93.43 94.55
ERFNet 86.42 92.71 91.10 94.38 87.17 93.14 92.31 93.99
UNet3+ 88.58 93.95 93.02 93.70 89.19 94.29 93.25 95.15
T-LinKNet 88.70 94.01 92.87 95.21 89.33 94.37 94.39 94.36
Res_ASPP_UNNet++ 88.35 93.81 94.00 94.18 88.91 94.13 93.65 94.61
DN-Net 89.20 94.29 94.31 94.27 89.85 94.65 94.12 95.20
Tab.7  Comparison of segmentation accuracy of different networks on the WHU dataset(%)
类型 原图 SegNet UNet ENet UNet++ ERFNet UNet3+ T-LinK
Net		 Res_
UNet++		 DN-Net 标签
小型建筑物
密集建筑物
大型建筑物
Tab.8  Visualization of different models on the WHU validation set
类型 原图 SegNet UNet ENet UNet++ ERFNet UNet3+ T-LinK
Net		 Res_
UNet++		 DN-Net 标签
小型建筑物
密集建筑物
大型建筑物
Tab.9  Visualization of different models on the WHU test set
Model FLOPs/109 参数量/106 IoU/% F1%
SegNet 0.5 0.4 88.07 93.66
UNet 31.1 13.4 88.24 93.75
ENet 0.5 0.4 86.62 92.83
UNet++ 17.5 24.7 88.65 93.99
ERFNet 3.4 2.1 87.17 93.14
UNet3+ 90.5 7.6 89.19 94.29
T-LinKNet 153.2 200.0 89.33 94.37
Res_ASPP_UNNet++ 8.8 4.5 88.91 94.13
DN-Net 43.2 12.9 89.85 94.65
Tab.10  Analysis of different network complexities in the WHU dataset
[1] Chen J, Xia M, Wang D, et al. Double branch parallel network for segmentation of buildings and waters in remote sensing images[J]. Remote Sensing, 2023, 15(6):1536.
doi: 10.3390/rs15061536 url: https://www.mdpi.com/2072-4292/15/6/1536
[2] Ji X, Tang L, Lu T, et al. DBENet:Dual-branch ensemble network for sea-land segmentation of remote-sensing images[J]. IEEE Transactions on Instrumentation and Measurement, 2023,72:5503611.
[3] Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
doi: 10.1109/TPAMI.2016.2572683 pmid: 27244717
[4] Ronneberger O, Fischer P, Brox T. U-net:Convolutional networks for biomedical image segmentation[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2015:234-241.
[5] Badrinarayanan V, Kendall A, Cipolla R. SegNet:A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12):2481-2495.
doi: 10.1109/TPAMI.2016.2644615 pmid: 28060704
[6] Noh H, Hong S, Han B. Learning deconvolution network for semantic segmentation[C]//2015 IEEE International Conference on Computer Vision (ICCV).December 7-13,2015, Santiago,Chile.IEEE, 2015:1520-1528.
[7] 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, 2018, 40(4):834-848.
doi: 10.1109/TPAMI.2017.2699184 url: http://ieeexplore.ieee.org/document/7913730/
[8] Chen LC, Papandreou G, Schroff F, et al. Rethinking atrous convolution for semantic image segmentation[J]. arXiv preprint arXiv:170605587, 2017.
[9] Chen L C, Zhu Y, Papandreou G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2018:833-851.
[10] Paszke A, Chaurasia A, Kim S, et al. ENet:A deep neural network architecture for real-time semantic segmentation[J/OL].(2016-06-07).https://arxiv.org/abs/1606.021470.
url: https://arxiv.org/abs/1606.02147
[11] Romera E, Álvarez J M, Bergasa L M, et al. ERFNet:Efficient residual factorized ConvNet for real-time semantic segmentation[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 19(1):263-272.
doi: 10.1109/TITS.2017.2750080 url: http://ieeexplore.ieee.org/document/8063438/
[12] Lin J, Jing W, Song H, et al. ESFNet:Efficient network for building extraction from high-resolution aerial images[J]. IEEE Access, 2822,7:54285-54294.
[13] Zhou Z W, Rahman Siddiquee MM, Tajbakhsh N, et al. Unet++:A nested u-net architecture for medical image segmentation[C]//Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support:4th International Workshop, DLMIA 2018,and 8th International Workshop,ML-CDS 2018,Held in Conjunction with MICCAI 2018,Granada,Spain,September 20,2018,Proceedings 4.Springer, 2018:3-11.
[14] Huang H, Lin L, Tong R, et al. UNet 3:A full-scale connected UNet for medical image segmentation[C]// ICASSP 2020-2020 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP). May 4-8,2020, Barcelona,Spain.IEEE, 2020:1055-1059.
[15] Liu Z, Lin Y T, Cao Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV).Montreal,QC,Canada. IEEE, 2021:9992-10002.
[16] 王振庆, 周艺, 王世新, 等. IEU-Net高分辨率遥感影像房屋建筑物提取[J]. 遥感学报, 2021, 25(11):2245-2254.
[16] Wang Z Q, Zhou Y, Wang S X, et al. House building extraction from high-resolution remote sensing images based on IEU-Net[J]. National Remote Sensing Bulletin, 2021, 25(11):2245-2254.
doi: 10.11834/jrs.20210042 url: http://www.ygxb.ac.cn/zh/article/doi/10.11834/jrs.20210042/
[17] 张立亭, 孔文学, 罗亦泳, 等. 改进LinkNet的高分辨率遥感影像建筑物提取方法[J]. 测绘科学, 2022, 47(9):120-127,145.
[17] Zhang L T, Kong W X, Luo Y Y, et al. Improved LinkNet building extraction method for high resolution remote sensing image[J]. Science of Surveying and Mapping, 2022, 47(9):120-127,145.
[18] 龙丽红, 朱宇霆, 闫敬文, 等. 新型语义分割D-UNet的建筑物提取[J]. 遥感学报, 2023, 27(11):2593-2602.
[18] Long L H, Zhu Y T, Yan J W, et al. New building extraction method based on semantic segmentation[J]. National Remote Sensing Bulletin, 2023, 27(11):2593-2602.
doi: 10.11834/jrs.20211029 url: http://www.ygxb.ac.cn/zh/article/doi/10.11834/jrs.20211029/
[19] 吕少云, 李佳田, 阿晓荟, 等. Res_ASPP_UNet++:结合分离卷积与空洞金字塔的遥感影像建筑物提取网络[J]. 遥感学报, 2023, 27(2):502-519.
[19] Lyu S Y, Li J T, A X H, et al. Res_ASPP_UNet++:Building an extraction network from remote sensing imagery combining depthwise separable convolution with atrous spatial pyramid pooling[J]. National Remote Sensing Bulletin, 2023, 27(2):502-519.
doi: 10.11834/jrs.20210477 url: http://www.ygxb.ac.cn/zh/article/doi/10.11834/jrs.20210477/
[20] 刘晨晨, 葛小三, 武永斌, 等. 基于混合注意力机制和Deeplabv3+的遥感影像建筑物提取方法[J]. 自然资源遥感, 2025, 37(1):31-37.doi:10.6046/zrzyyg.2023295.
[20] Liu C C, Ge X S, Wu Y B, et al. A method for information extraction of buildings from remote sensing image based on hybrid attention mechanism and Deeplabv3+[J] Remote Sensing for Natural Resource, 2025, 37(1):31-37.doi:10.6046/zrzyyg.2023295.
[21] 季顺平, 魏世清. 遥感影像建筑物提取的卷积神经元网络与开源数据集方法[J]. 测绘学报, 2019, 48(4):448-459.
doi: 10.11947/j.AGCS.2019.20180206
[21] Ji S P, Wei S Q. Building extraction via convolutional neural networks from an open remote sensing building dataset[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(4):448-459.
[22] Lee C Y, Xie S, Gallagher P, et al. Deeply-supervised nets[C]// Artificial Intelligence and Statistics. PMLR, 2015:562-570.
[1] ZHU Juanjuan, HUANG Liang, ZHU Shasha. SD-BASNet：a building extraction network for high-spatial-resolution remote sensing imagery[J]. Remote Sensing for Natural Resources, 2025, 37(5): 122-130.
[2] LIU Chenchen, GE Xiaosan, WU Yongbin, YU Haikun, ZHANG Beibei. A method for information extraction of buildings from remote sensing images based on hybrid attention mechanism and Deeplabv3+[J]. Remote Sensing for Natural Resources, 2025, 37(1): 31-37.
[3] WU Yu, ZHANG Jun, LI Yixu, HUANG Kangyu. Research on building cluster identification based on improved U-Net[J]. Remote Sensing for Land & Resources, 2021, 33(2): 48-54.
[4] LU Qi, QIN Jun, YAO Xuedong, WU Yanlan, ZHU Haochen. Buildings extraction of GF-2 remote sensing image based on multi-layer perception network[J]. Remote Sensing for Land & Resources, 2021, 33(2): 75-84.
[5] CUI Shiai, CHENG Bo, LIU Yueming. Research on methods of building area extraction from high resolution SAR image based on manifold learning[J]. REMOTE SENSING FOR LAND & RESOURCES, 2017, 29(4): 48-56.
[6] FU Ying, GUO Qiaozhen, PAN Yingyang, WANG Dongchuan. Research on building extraction rules based on SPOT6 data[J]. REMOTE SENSING FOR LAND & RESOURCES, 2017, 29(3): 65-69.
[7] WANG Xudong, DUAN Fuzhou, QU Xinyuan, LI Dan, YU Panfeng. Building extraction based on UAV imagery data with the synergistic use of objected-based method and SVM classifier[J]. REMOTE SENSING FOR LAND & RESOURCES, 2017, 29(1): 97-103.
[8] WANG Xue, LI Peijun, JIANG Shasha, LIU Jing, SONG Benqin. Building extraction using airborne LiDAR data and very high resolution imagery over a complex urban area[J]. REMOTE SENSING FOR LAND & RESOURCES, 2016, 28(2): 106-111.
[9] LI Feng, CUI Ximin, YUAN Debao, WANG Qiang, WU Yajun. Extraction of the building region from airborne LiDAR point clouds under complex urban conditions[J]. REMOTE SENSING FOR LAND & RESOURCES, 2013, 25(3): 85-89.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
京ICP备05055290号-2
Copyright © 2017 Remote Sensing for Natural Resources
Support by Beijing Magtech