| 482 | 5 | 259 |
| 下载次数 | 被引频次 | 阅读次数 |
针对高分辨率遥感影像建筑物提取精度不高、易出现误提和漏提等问题，提出了一种改进U-Net网络的建筑物提取方法。以Res Net50为U-Net模型的编码器部分，同时引入CBAM混合注意力机制和FPN特征金字塔结构对网络进行优化，从而提高网络提取建筑物信息的准确度和稳健性。基于高景一号遥感影像，制作512×512大小的样本进行训练，并与U-Net、基于Res Net50骨干网络的U-Net网络和Deep Labv3+进行对比验证。结果表明，该算法具有更强的分割效果和性能，适用于不同类型的高分辨率建筑物提取任务。
Abstract：In view of the low accuracy and misrepresentation and omission of building extraction in high-resolution remote sensing images, we proposed a building extraction method based on improved U-Net network. We used Res Net50 as the encoder part of U-Net model, and introduced CBAM hybrid attention mechanism and FPN feature pyramid structure to optimize the network to improve the accuracy and robustness of network in building information extraction. Based on Super View-1 remote sensing images, we produced 512 × 512 samples for training, and compared with U-Net, the U-Net network based on Res Net50 backbone network and Deep Labv3+. The results show that this method proposed in this article has a stronger segmentation effect and performance, and can apply to different types of high-resolution building extraction tasks.
[1]魏麟.基于高分辨率遥感影像的震灾建筑物损毁检测[J].地理空间信息，2022,20(3):68-71
[2]陆妍如，毛辉辉，贺琰，等.基于Deeplab v3+的高分辨率遥感影像地物分类研究[J].地理空间信息，2022,20(6):1-6
[3]许开宏，施招，马磊超，等.基于机载激光雷达与高景一号数据的草原地上生物量反演研究[J].草业学报，2023,32(5):40-49
[4] Huang X,Zhang L. An SVM Ensemble Approach Combining Spectral,Structural,and Semantic Features for the Classification of High-resolution Remotely Sensed Imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(1):257-272
[5]高贤君，郑学东，沈大江，等.城郊高分影像中利用阴影的建筑物自动提取[J].武汉大学学报(信息科学版)，2017,42(10):1 350-1 357
[6]杨栋杰，高贤君，冉树浩，等.基于多重多尺度融合注意力网络的建筑物提取[J].浙江大学学报(工学版)，2022,56(10):1 924-1 934
[7]金澍，关沫，边玉婵，等.基于改进U-Net的遥感影像建筑物提取方法[J].激光与光电子学进展，2023,60(4):59-65
[8]秦梦宇，刘勇，张寅丹，等.基于改进U-Net模型的高分辨率遥感影像中城市建筑物的提取[J].兰州大学学报(自然科学版)，2022,58(2):254-261
[9]陈嘉浩，邢汉发，陈相龙.融合级联CRFs和U-Net深度学习模型的遥感影像建筑物自动提取[J].华南师范大学学报(自然科学版)，2022,54(1):70-78
[10]卢彻，徐胜华，朱军.改进U-Net的高分影像建筑物提取方法[J].测绘科学，2021,46(12):140-146
[11] Ronneberger O,Fischer P,Brox T. U-Net:Convolutional Networks for Biomedical Image Segmentation[C]. International Conference on Medical Image Computing and Computer-assisted Intervention,2015:234-241
[12] He K,Zhang X,Ren S,et al. Deep Residual Learning for Image Recognition[C]. IEEE Conference on Computer Vision and Pattern Recognition(CVPR),Las Vegas,2016:770-778
[13] 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
[14] Lin T Y,Dollár P,Girshick R,et al. Feature Pyramid Networks for Object Detection[C]. IEEE Conference on Computer Vision and Pattern Recognition,2017:936-944
[15] 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
[16]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
基本信息:
中图分类号:P237
引用信息:
[1]陶从辉,高青山,赵梦琳.改进U-Net网络的高分辨率遥感影像建筑物提取方法[J].地理空间信息,2025,23(01):9-12.
基金信息:
浙江省生态环境科研和成果推广项目（2021HT0061）
2025-01-24
2025-01-24