全卷积神经网络用于遥感影像水体提取

doi:10.13474/j.cnki.11-2246.2018.0173

测绘通报 ›› 2018, Vol. 0 ›› Issue (6): 41-45.doi: 10.13474/j.cnki.11-2246.2018.0173

全卷积神经网络用于遥感影像水体提取

王雪¹, 隋立春^1,2, 钟棉卿¹, 李顶萌³, 党丽丽⁴

1. 长安大学, 陕西西安 710054;
2. 地理国情监测国家测绘地理信息局工程技术研究中心, 陕西西安 710054;
3. 西安国际港务区规划局, 陕西西安 710026;
4. 咸阳职业技术学院, 陕西咸阳 712000

收稿日期:2017-10-11 修回日期:2018-03-27 出版日期:2018-06-25 发布日期:2018-07-07
作者简介:王雪(1983-),女,博士生,讲师,主要研究方向为遥感数字图像处理与影像分析。E-mail:wangxue-2001@163.com
基金资助:
国家自然科学基金（41372330）；国家自然科学基金青年基金（41601345）

Fully Convolution Neural Networks for Water Extraction of Remote Sensing Images

WANG Xue¹, SUI Lichun^1,2, ZHONG Mianqing¹, LI Dingmeng³, DANG Lili⁴

1. College of Geology Engineering and Geomatics, Chang'an University, Xi'an 710054, China;
2. Engineering Research Center, Geographical Conditions Monitoring National Administration of Surveying, Mapping and Geoinformation, Xi'an 710054, China;
3. Planning Bureau, Xi'an International Trade and Logistics Park, Xi'an 710026, China;
4. Xianyang Vocational Technical College, Xianyang 712000, China

Received:2017-10-11 Revised:2018-03-27 Online:2018-06-25 Published:2018-07-07

摘要/Abstract

摘要： 提出了一种全卷积神经网络模型用于遥感影像的水体目标提取的方法，介绍了全卷积神经网络的基本原理及构建3种网络模型的过程。为了分析模型效果，首先搜集带有水体的影像数据，并将这些数据标注为水体和背景两类，然后利用构建的全卷积神经网络模型进行推理和学习获取先验模型，最后对测试影像进行水体提取试验。通过与传统的基于影像光谱特征的阈值法和基于图论的GrabCut算法提取结果进行比较，验证了本文方法的可行性。

关键词: 遥感影像, 水体提取, 全卷积神经网络, 阈值法, GrabCut算法

Abstract: This paper presented fully convolution neural networks for extracting water from remote sensing images.The paper introduced the basic principle and method of three kinds of models of fully convolution neural networks.For extracting the water target,firstly,the image data were gathered and labeled two classes——water and background.By inference and learning in fully convolution neural networks,the trained models were obtained.Finally,water target of the test images was extracted.The feasibility of the proposed models was validated by comparing the result of extraction with that of the traditional thresholding method based on image spectral features and GrabCut algorithm based on graph theory.

Key words: remote sensing images, water extraction, fully convolution neural networks, thresholding method, GrabCut

中图分类号:

P237

王雪, 隋立春, 钟棉卿, 李顶萌, 党丽丽. 全卷积神经网络用于遥感影像水体提取[J]. 测绘通报, 2018, 0(6): 41-45.

WANG Xue, SUI Lichun, ZHONG Mianqing, LI Dingmeng, DANG Lili. Fully Convolution Neural Networks for Water Extraction of Remote Sensing Images[J]. 测绘通报, 2018, 0(6): 41-45.

参考文献

[1] 毕海芸,王思远,曾江源,等.基于TM影像的几种常用水体提取方法的比较和分析[J].遥感信息,2012,27(5):77-82.
[2] 李小涛,黄诗峰,郭怀轩.基于纹理特征的SPOT5影像水体体取方法研究[J].人民黄河,2010,32(12):5-6.
[3] 殷亚秋,李家国,余涛,等.基于高分辨率遥感影像的面向对象水体提取方法研究[J].测绘通报,2015(1):81-85.
[4] 吴岸城.神经网络与深度学习[M].北京:电子工业出版社,2016.
[5] KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet Classification with Deep Convolutional Neural Networks[C]//International Conference on Neural Information Processing Systems.Nevada:Curran Associates Inc,2012:1097-1105.
[6] SIMONYAN K,ZISSERMAN A.Very Deep Convolutional Networks for Large-scale Image Recognition[J].Computer Science,2014(9):1409-1456.
[7] SZEGEDY C,LIU W,JIA Y,et al.Going Deeper with Convolutions[C]//IEEE Conference on Computer Vision and Pattern Recognition.Santiago:IEEE Computer Society,2015:1-9.
[8] GIRSHICK R,DONAHUE J,DARREL T,et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[J].Computer Science,2013(6):580-587.
[9] GIRSHICK R.Fast R-CNN[C]//International Conference on Computer Vision.Santiago:IEEE,2015:1440-1448.
[10] REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards Real-time Object Detection with Region Proposal Networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.
[11] HE K,ZHANG X,REN S,et al.Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1904-1916.
[12] LONG J,SHELHAMER E,DARREL T.Fully Convolutional Networks for Semantic Segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(4):640-651.
[13] DUMOULIN V,VISIN F.A Guide to Convolution Arithmetic for Deep Learning[J].Stat,2016,1050:23.
[14] ZEILER M D,FERGUS R.Visualizing and Understanding Convolutional Networks[C]//European Conference on Computer Vision.Zurich:Springer, 2014:818-833.
[15] ROTHER C,KOLMOGOROV V,BLAKE A."GrabCut":Interactive Foreground Extraction Using Iterated Graph Cuts[J].IEEE Transactions on Graphics,2004,23(3):309-314.
[16] GARCIAGARCIA A,ORTSESCOLANO S,OPREA S,et al.A Review on Deep Learning Techniques Applied to Semantic Segmentation[J].Computer Science,2017(4):1704.

全卷积神经网络用于遥感影像水体提取

Fully Convolution Neural Networks for Water Extraction of Remote Sensing Images

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李茂林, 闫庆武, 仲晓雅. 一种复合型校正植被信息的方法[J]. 测绘通报, 2022, 0(6): 49-54142.
[2]	任超, 邓诗琴, 高懋芳. 一种提取北部湾沿海地区水体信息的动态阈值方法[J]. 测绘通报, 2022, 0(5): 14-19,37.
[3]	王春霞, 张俊, 李屹旭, 范成成. 一种基于Landsat 8的多波段组合水体指数模型[J]. 测绘通报, 2022, 0(5): 20-25.
[4]	张龚泉, 岳建平, 刘胜男, 王海龙, 杨智翔, 周航宇. 基于遥感影像的鄱阳湖地形提取方法[J]. 测绘通报, 2022, 0(5): 62-66,73.
[5]	潘建平, 付占宝, 程卫华, 陆世安, 李鑫, 孙博文, 黄桂萍. 基于地表类型的山地海绵城市生态建设评估[J]. 测绘通报, 2022, 0(5): 84-88.
[6]	陈智朗, 付振华, 朱紫阳, 王慧慧, 刘沁雯, 杨钰灵, 许耿然. 基于HRNet的高分辨率遥感影像建筑物变化信息提取[J]. 测绘通报, 2022, 0(5): 126-132.
[7]	刘小玉, 刘扬, 杜明义, 张敏, 贾竞珏, 杨恒. 基于DeeplabV3+的建筑垃圾堆放点识别[J]. 测绘通报, 2022, 0(4): 16-19,43.
[8]	王雪, 梁珂, 隋立春, 钟棉卿, 朱剑锋. 膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取[J]. 测绘通报, 2022, 0(4): 61-65.
[9]	高鸣, 周鑫鑫, 刘琦, 杨光迪, 吴长彬. 深度学习网络支持下的农房侵占耕地自动化监测[J]. 测绘通报, 2022, 0(3): 47-53.
[10]	王胜利, 朱寿红, 蒋毅. 基于CNN模型迁移的OLI影像光伏电池板场景识别[J]. 测绘通报, 2022, 0(2): 5-9.
[11]	范媛, 范宏, 吴建国, 林峻, 陈吉军, 巴日斯, 郑江华. 无人机影像识别白喉乌头的相对高程阈值法适用性分析[J]. 测绘通报, 2022, 0(2): 131-135.
[12]	杨素妨, 杜林. 多特征流形鉴别嵌入的高分辨率遥感影像分类[J]. 测绘通报, 2021, 0(9): 37-42.
[13]	耿万轩, 周维勋, 金双根. 基于LDCNN特征提取的多核SVM高分辨率遥感影像场景分类[J]. 测绘通报, 2021, 0(8): 14-21.
[14]	苏艺凡, 党建武, 王阳萍, 杨景玉. 改进的区间二型模糊聚类遥感影像变化检测[J]. 测绘通报, 2021, 0(7): 44-51,58.
[15]	聂倩, 七珂珂, 赵艳福. 融入超像素分割的高分辨率影像面向对象分类[J]. 测绘通报, 2021, 0(6): 44-49.