一种深度学习的无人机影像道路自动提取方法
An automatic road extraction method from UAV imagery based on deep learning
查看参考文献19篇
|
文摘
|
针对无人机影像道路提取自动化程度低、道路信息不完整及道路交叉口不连通等问题,该文提出了一种结合拓扑结构和全局上下文感知的无人机影像道路提取方法,通过构建一种编码/解码模式的深度学习方法实现自动化提取。在网络模型中,设计了聚合特征模块及增强型扩张卷积模块以获取更多的道路信息,并引入拓扑感知损失函数以保证道路的连通性,实现道路拓扑结构特性的反演。实验结果表明,基于改进后的网络模型对道路信息的提取效果较好,在无人机影像测试集上的准确率、召回率、F1得分和交并比(IoU)分别达到了89.07%、84.74%、 86.86%和72.45%;在马萨诸塞州道路公共影像集通用性测试中,提取原始遥感图像的道路信息也表现了出色的提取性能。 |
|
其他语种文摘
|
For the problem of the low degree of automation of road extraction from UAV images, incomplete road information and disconnection of road intersections,this paper designs a road extraction method from UAV images that combines topology and global a UAV image road extraction method com bining topological structure and global context-aw are.This method constructs a deep learning method of encoding/decoding mode to realize automatic extraction.In the network model,we designed an aggregate feature module and an enhanced expanded convolution module to obtain more road information.Considering the topological structure of the road,a topology-aware loss function is introduced to ensure the connectivity of the road.The experimental results show that the extraction effect of road information based on the improved network model in this paper is better. Respectively,the accuracy,recall,F1-score and IoU on the UAV image set reached 89.07%、84.74%、86.86%and 72.45%.In addition,in the general test of the public road image set in Massachusetts,the road information extracted from the original remote sensing image also showed excellent extraction performance. |
|
来源
|
测绘科学
,2021,46(11):106-113 【核心库】
|
|
DOI
|
10.16251/j.cnki.1009-2307.2021.11.015
|
|
关键词
|
拓扑结构
;
全局上下文
;
无人机
;
道路提取
;
聚合特征
|
|
地址
|
1.
中国科学院地理科学与资源研究所, 资源与环境信息系统国家重点实验室, 北京, 100101
2.
长江大学地球科学学院, 武汉, 430100
3.
天津中科无人机应用研究院, 天津, 301800
4.
中国科学院无人机应用与管控研究中心, 北京, 100101
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
1009-2307 |
|
学科
|
测绘学 |
|
基金
|
国家自然科学基金
;
中国科学院重点部署项目
;
天津科技计划项目智能制造专项
|
|
文献收藏号
|
CSCD:7097058
|
参考文献 共
19
共1页
|
|
1.
廖小罕. 地理科学发展与新技术应用.
地理科学进展,2020,39(5):709-715
|
CSCD被引
23
次
|
|
|
|
|
2.
Zhong Z L. Fully convolutional networks for building and road extraction:Preliminary results.
2016IEEE International Geoscience and Remote Sensing Symposium(IGARSS),2016:1591-1594
|
CSCD被引
1
次
|
|
|
|
|
3.
Chen K Q. Semantic segmentation of aerial images with shuffling convolutional neural networks.
IEEE Geoscience and Remote Sensing Letters,2018,15(2):173-177
|
CSCD被引
9
次
|
|
|
|
|
4.
王龙飞. 道路交叉口自动检测算法的研究.
测绘科学,2020,45(5):126-131
|
CSCD被引
4
次
|
|
|
|
|
5.
郭正胜. U型卷积神经网络的ZY-3影像道路提取方法.
测绘科学,2020,45(4):51-57
|
CSCD被引
8
次
|
|
|
|
|
6.
Mnih V. Learning to detect roads in high-resolution aerial images.
Computer Vision-ECCV,2010:210-223
|
CSCD被引
1
次
|
|
|
|
|
7.
Li P K. Road network extraction via deep learning and line integral convolution.
2016IEEE International Geoscience and Remote Sensing Symposium(IGARSS),2016:1599-1602
|
CSCD被引
1
次
|
|
|
|
|
8.
Xu Y Y. Road extraction from high-resolution remote sensing imagery using deep learning.
Remote Sensing,2018,10(9):1461
|
CSCD被引
17
次
|
|
|
|
|
9.
Wang S. An improved method for road extraction from high-resolution remotesensing images that enhances boundary information.
Sensors,2020,20(7):2064
|
CSCD被引
7
次
|
|
|
|
|
10.
Ronneberger O. U-Net: convolutional networks for biomedical image segmentation.
Medical Image Computing and Computer-Assisted Intervention,2015:234-241
|
CSCD被引
93
次
|
|
|
|
|
11.
He K M. Identity mappings in deep residual networks.
European Conference on Computer Vision,2016:630-645
|
CSCD被引
33
次
|
|
|
|
|
12.
Zhao H S. Pyramid scene parsing network.
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2017:2881-2890
|
CSCD被引
3
次
|
|
|
|
|
13.
Chen L C. DeepLab:semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected CRFs.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848
|
CSCD被引
1332
次
|
|
|
|
|
14.
Tao C. Spatial information inference net:road extraction using road-specific contextual information.
ISPRS Journal of Photogrammetry and Remote Sensing,2019,158(12):155-166
|
CSCD被引
2
次
|
|
|
|
|
15.
Neven D. Towards end-to-end lane detection:an instance segmentation approach.
IEEE International Geoscience and Remote Sensing Symposium(IGARSS),2018:286-291
|
CSCD被引
1
次
|
|
|
|
|
16.
He H. Road extraction by using atrous spatial pyramid pooling integrated encoder-decoder network and structural similarity loss.
Remote Sensing,2019,11(9):1015
|
CSCD被引
12
次
|
|
|
|
|
17.
Liu H F. Deep representation learning for road detection using Siamese network.
Multimedia Tools and Applications,2019,78(17):269-283
|
CSCD被引
3
次
|
|
|
|
|
18.
Mosinska A. Beyond the pixel-wise loss for topologyaware delineation.
IEEE/CVF Conference on Computer Vision and Pattern Recognition,2018:3136-3145
|
CSCD被引
1
次
|
|
|
|
|
19.
Zhou L C. D-LinkNet:linknet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction.
2018IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops(CVPRW),2018:192-1924
|
CSCD被引
1
次
|
|
|
|
|
了解气象卫星更多信息，请访问