基于Deeplabv3+与CRF的遥感影像典型要素提取方法

doi:10.19678/j.issn.1000-3428.0053359

计算机工程 ›› 2019, Vol. 45 ›› Issue (10): 260-265,271. doi: 10.19678/j.issn.1000-3428.0053359

基于Deeplabv3+与CRF的遥感影像典型要素提取方法

王俊强^1,2, 李建胜¹, 周华春², 张旭¹

1. 信息工程大学地理空间信息学院, 郑州 450000;
2. 78123部队, 成都 610000

收稿日期:2018-12-10 修回日期:2019-01-17 出版日期:2019-10-15 发布日期:2019-01-23
作者简介:王俊强(1990-),男,硕士研究生,主研方向为计算机视觉、导航定位与服务;李建胜(通信作者),教授;周华春,工程师;张旭,硕士研究生。
基金资助:
国家自然科学基金（41876150）。

Typical Element Extraction Method of Remote Sensing Image Based on Deeplabv3+ and CRF

WANG Junqiang^1,2, LI Jiansheng¹, ZHOU Huachun², ZHANG Xu¹

1. Institute of Geospatial Information, Information Engineering University, Zhengzhou 450000, China;
2. 78123 Troops, Chengdu 610000, China

Received:2018-12-10 Revised:2019-01-17 Online:2019-10-15 Published:2019-01-23

摘要/Abstract

摘要： 为提取高分辨率遥感影像的典型要素（建筑物及道路），基于深度学习，提出一种语义分割与全连接条件随机场（CRF）相结合的提取方法。以Deeplabv3+作为语义分割模型，提取较完整图像分割信息，并将其作为全连接CRF的一元能量函数的输入，利用平均场近似方法进行推理，实现对分割信息边界的优化。通过分析Deeplabv3+模型在噪声样本集数据的训练效果验证其鲁棒性，并基于公开影像及矢量数据源设计大规模遥感训练样本集智能采集系统。采集罗德岛2 000平方公里遥感影像及相对应典型要素标记数据作为样本进行实验，结果表明，该方法分割精度MIoU值达到80.32%，结合形态学滤波处理，要素边界轮廓明显优于初始分割结果。

关键词: 遥感影像, 语义分割, 条件随机场, 要素提取, 深度学习

Abstract: In order to extract the typical elements (buildings and roads) of high-resolution remote sensing images,based on deep learning,a extraction method combining semantic segmentation and full-connection Conditional Random Field (CRF) is proposed.Using Deeplabv3+ as the semantic segmentation model,the relatively complete image segmentation information is extracted and used as the input of the unary potential function of the fully connected CRF.The mean field approximation method is used to realize the optimization of the segmentation information boundary.The robustness of Deeplabv3+ model is verified by analyzing the training effect of Deeplabv3+ in noise sample set data and based on public image and vector data source,a large-scale remote sensing training sample set intelligent collection system is designed.Collecting 2 000 square kilometers of remote sensing images of Rhode Island and taking corresponding typical elemental marker data as samples to conduct experiments,the results show that the MIoU value of the proposed method reaches 80.23%.After morphological processing,the boundary of the element is obviously clearer than the intial segmentation raslut.

Key words: remote sensing image, semantic segmentation, Conditional Random Field(CRF), element extraction, deep learning

中图分类号:

TP751

王俊强, 李建胜, 周华春, 张旭. 基于Deeplabv3+与CRF的遥感影像典型要素提取方法[J]. 计算机工程, 2019, 45(10): 260-265,271.

WANG Junqiang, LI Jiansheng, ZHOU Huachun, ZHANG Xu. Typical Element Extraction Method of Remote Sensing Image Based on Deeplabv3+ and CRF[J]. Computer Engineering, 2019, 45(10): 260-265,271.

http://www.ecice06.com/CN/Y2019/V45/I10/260

图/表 11

20191014195701

20191014195705

20191014195709

20191014195711

20191014195714

20191014195717

20191014195720

20191014195724

20191014195728

20191014195732

20191014195735

参考文献 21

[1]	蔺海峰,马宇峰,宋涛.基于SIFT特征目标跟踪算法研究[J].自动化学报,2010,36(8):1204-1208.
[2]	NG H P,ONG S H,FOONG K W C,et al.Medical image segmentation using k-means clustering and improved watershed algorithm[C]//Proceedings of Image Analysis and Interpretatio.Southwest Symposium.Washington D.C.,USA:IEEE Press,2006:61-65.
[3]	陶超,谭毅华,蔡华杰,等.面向对象的高分辨率遥感影像城区建筑物分级提取方法[J].测绘学报,2010,39(1):39-45.
[4]	KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Image net classification with deep convolutional neural networks[C]//Proceedings of Advances in Neural Information Processing Systems.Lake Tahoe,USA:NIPS,2012:1097-1105.
[5]	SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[EB/OL].[2018-12-01].https://arxiv.org/abs/1409.1556.
[6]	SZEGEDY C,LIU Wei,JIA Yangqing,et al.Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:1-9.
[7]	LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:3431-3440.
[8]	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.
[9]	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.
[10]	LIN Guosheng,MILAN A,SHEN Chuanhua,et al.RefineNet:multi-path refinement networks for high-resolution semantic segmentation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:5168-5177.
[11]	CHEN L C,ZHU Yunkun,PAPANDREOU G,et al.Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of ECCV’18.Berlin,Germany:Springer,2018:801-818.
[12]	冯丽英.基于深度学习技术的高分辨率遥感影像建设用地信息提取研究[D].杭州:浙江大学,2017.
[13]	陈洋,范荣双,王竞雪,等.基于深度学习的资源三号卫星遥感影像云检测方法[J].光学学报,2018,38(1):362-367.
[14]	CHENG Guangliang,WANG Ying,XU Shibiao,et al.Automatic road detection and centerline extraction via cascaded end-to-end convolutional neural network[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(6):3322-3337.
[15]	KRAHENBUHL P,KOLTUN V.Efficient inference in fully connected CRFs with Gaussian edge potentials[J].Advances in Neural Information Processing Systems,2011,24:109-117.
[16]	CHOLLET F.Xception:deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:1251-1258.
[17]	HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:770-778.
[18]	MNIH V.Machine learning for aerial image labeling[D].Toronto,Canada:University of Toronto,2013.
[19]	HAKLAY M,WEBER P.Openstreetmap:User-generated street maps[J].IEEE Pervas Computing,2008,7(4):12-18.
[20]	胡正平,杨建秀.HOG特征混合模型结合隐SVM的感兴趣目标检测定位算法[J].信号处理,2011,27(8):1206-1212.
[21]	DE BOER P T,KROESE D P,MANNOR S,et al.A tutorial on the cross-entropy method[J].Annals of Operations Research,2005,134(1):19-67.

选择文件类型/文献管理软件名称

选择包含的内容

基于Deeplabv3+与CRF的遥感影像典型要素提取方法

Typical Element Extraction Method of Remote Sensing Image Based on Deeplabv3+ and CRF

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价

[1]	徐春波, 闫娟, 杨慧斌, 王博, 吴晗. 基于目标检测和语义分割的视觉SLAM算法[J]. 计算机工程, 2023, 49(8): 199-206, 214.
[2]	江雨燕, 陶承凤, 李平. 数据增强和自适应自步学习的深度子空间聚类算法[J]. 计算机工程, 2023, 49(8): 96-103, 110.
[3]	李泽水, 冀俊忠, 杨翠翠. 基于边权重信息深度网络嵌入的PPIN功能模块检测[J]. 计算机工程, 2023, 49(8): 69-76.
[4]	王可铮, 徐玉芬, 周尚波. 结合对比感知损失和融合注意力的图像去雾模型[J]. 计算机工程, 2023, 49(8): 207-214.
[5]	刘俊豪, 王美林, 谢兴, 宋烨兴, 许莉花. 基于改进YOLOv5的皮革瑕疵检测算法[J]. 计算机工程, 2023, 49(8): 240-249.
[6]	闫兴亚, 匡娅茜, 白光睿, 李月. 基于深度学习的学生课堂行为识别方法[J]. 计算机工程, 2023, 49(7): 251-258.
[7]	李军侠, 王星驰, 殷梓, 石德硕. 边缘深度挖掘的弱监督显著性目标检测[J]. 计算机工程, 2023, 49(7): 169-178.
[8]	吴珊, 周凤. 基于改进SSD算法的小目标检测[J]. 计算机工程, 2023, 49(7): 179-188.
[9]	席建锐, 唐红梅, 梁春阳, 刘鑫. 基于改进隐函数的点云物体重建[J]. 计算机工程, 2023, 49(7): 214-222.
[10]	齐咏生, 杜晓旭, 朱俊峰, 高胜利, 刘利强. 基于增强型轻量深度网络的牧区牲畜高效检测[J]. 计算机工程, 2023, 49(7): 278-287.
[11]	谌雨章, 黄逸姿, 张钧涵. 基于多速率空洞卷积的多尺度水下小目标检测[J]. 计算机工程, 2023, 49(6): 257-264.
[12]	张博旭, 蒲智, 程曦. 基于提示学习的维吾尔语文本分类研究[J]. 计算机工程, 2023, 49(6): 292-299,313.
[13]	于海洋, 景鹏, 张文涛, 谢赛飞, 滑志华, 宋草原. 基于残差与注意力机制的道路裂缝检测U-Net改进模型[J]. 计算机工程, 2023, 49(6): 265-273.
[14]	王爱玲, 马文臻, 邹自明, 钟佳. 基于领域自适应的卫星工程参数异常检测[J]. 计算机工程, 2023, 49(5): 29-37,47.
[15]	宋羽凯, 谢江. 基于多任务学习的轻量级语音情感识别模型[J]. 计算机工程, 2023, 49(5): 122-128.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于Deeplabv3+与CRF的遥感影像典型要素提取方法

Typical Element Extraction Method of Remote Sensing Image Based on Deeplabv3+ and CRF

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价