| 结合深度学习和图割法的遥感影像建筑物检测 |
| |
| 引用本文： | 刘舸,邓兴升.结合深度学习和图割法的遥感影像建筑物检测[J].测绘通报,2019,0(11):69-73. |
| |
| 作者姓名： | 刘舸 邓兴升 |
| |
| 作者单位： | 长沙理工大学交通运输工程学院,湖南 长沙,410114;长沙理工大学交通运输工程学院,湖南 长沙,410114 |
| |
| 基金项目： | 湖南省教育厅资助科研项目（17B004） |
| |
| 摘 要： | 提出一种基于卷积神经网络和图割法的自动提取高分影像建筑物的方法。首先，通过卷积神经网络定位与检测建筑物的位置，逐一提取单个建筑物轮廓，利用检测结果分别建立建筑物和非建筑物的高斯混合模型（GMM），然后结合最大流最小割的图像分割方式实现全局优化，完成建筑物初步提取，最后用形态学进行优化。通过试验证明了该方法的可行性。
|
| 关 键 词： | 高分辨率正射图像 深度学习 建筑物信息提取 图割 卷积神经网络 |
| 收稿时间： | 2019-03-25 |
| 修稿时间： | 2019-05-23 |
Remote sensing image building extraction based on deep learning and graph cut |
| |
| LIU Ge,DENG Xingsheng.Remote sensing image building extraction based on deep learning and graph cut[J].Bulletin of Surveying and Mapping,2019,0(11):69-73. |
| |
| Authors: | LIU Ge DENG Xingsheng |
| |
| Institution: | School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha 410114, China |
| |
| Abstract: | A method for automatically extracting buildings on high-resolution images based on convolutional neural networks and graph cuts is proposed. Firstly, the location of the contour of the building is located and detected by the convolutional neural network, and the outlines of the individual buildings are extracted one by one. The Gaussian mixture model (GMM) of the building and the non-building is respectively established by the detection result, and the minimum flow is minimized. The cut image segmentation method achieves global optimization, completes the preliminary extraction of the building, and finally optimizes with morphology. The feasibility of the method is proved by experiments. |
| |
| Keywords: | high-resolution remote sensing image deep learning building information extraction graph cuts convolutional neural networks |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
