| 高分辨率遥感影像建筑物提取多路径RSU网络法 |
| |
| 引用本文： | 张玉鑫,颜青松,邓非.高分辨率遥感影像建筑物提取多路径RSU网络法[J].测绘学报,2022,51(1):135-144. |
| |
| 作者姓名： | 张玉鑫 颜青松 邓非 |
| |
| 作者单位： | 武汉大学测绘学院, 湖北 武汉 430079 |
| |
| 基金项目： | 四川省科技计划(2019YFG0460)~~； |
| |
| 摘 要： | 针对卷积神经网络在提取建筑物的过程中，存在建筑物边界不准确和建筑物内部空洞等问题，提出以RSU模块（residual U-block）为核心的MPRSU-Net （multi-path residual U-block network）。该模块利用编码器-解码器结构和残差连接，实现了局部特征和多尺度特征的融合。由于一个RSU模块提取的信息有限，MPRSU-Net进一步通过多路径结构并行了不同尺度的RSU模块，并在这些模块之间进行信息交换，提高了特征聚集效率。在分辨率为0.3 m的WHU和Inria建筑物数据集上进行试验，精度分别达95.65%和88.63%，IoU分别达91.17%和79.31%，验证了本文方法的有效性。此外，本文方法相较于U2Net，计算量明显降低，模型参数量减少68.63%，表明本文方法具有一定的应用价值。
|
| 关 键 词： | 高分辨率遥感影像 建筑物提取 多尺度 卷积神经网络 多路径 |
| 收稿时间： | 2020-10-14 |
| 修稿时间： | 2021-07-21 |
Multi-path RSU network method for high-resolution remote sensing image building extraction |
| |
| ZHANG Yuxin,YAN Qingsong,DENG Fei.Multi-path RSU network method for high-resolution remote sensing image building extraction[J].Acta Geodaetica et Cartographica Sinica,2022,51(1):135-144. |
| |
| Authors: | ZHANG Yuxin YAN Qingsong DENG Fei |
| |
| Institution: | School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China |
| |
| Abstract: | Inaccurate boundaries and holes are two major problems when extracting buildings from high-resolution remote sensing images by a convolution network.To solve these problems,we proposed the MPRSU-Net(multi-path residual U-block network),which is based on the RSU(residual U-block).The RSU is able to fuse local features and multi-scale features,with the help of the encoder-decoder structure and the residual connection.However,a single RSU is not enough to gather enough information,MPRSU-Net parallels RSU blocks of different scales by the multi-path structure and exchanges information among these blocks to further enhance the feature aggregation efficiency.Experimental results showed that the MPRSU-Net achieved 95.65%,88.63%precision,and 91.17%,79.31%IoU on 0.3 m resolution WHU and Inria building datasets,which showed the effectiveness of the proposed method.In addition,compared with the U2Net,MPRSU-Net is much lighter in computation and reduces the amount of model parameters by 68.63%,demonstrating that the method has some application value. |
| |
| Keywords: | high-resolution remote sensing image building extraction multi-scale convolutional neural networks multi-path |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《测绘学报》浏览原始摘要信息 |
| 点击此处可从《测绘学报》下载免费的PDF全文 |
|
