基于三维卷积神经网络的湖泊提取算法

doi:10.11772/j.issn.1001-9081.2019081436

计算机应用 ›› 2019, Vol. 39 ›› Issue (12): 3450-3455.DOI: 10.11772/j.issn.1001-9081.2019081436

• 第十七届中国机器学习会议(CCML 2019)论文 • 上一篇下一篇

基于三维卷积神经网络的湖泊提取算法

徐姗姗¹, 颜超², 高琳明¹

1. 南京林业大学信息科学技术学院, 南京 210037;
2. 国防科技大学气象海洋学院, 南京 211101

收稿日期:2019-04-29 修回日期:2019-08-15 出版日期:2019-12-10 发布日期:2019-09-04
作者简介:徐姗姗(1988-),女,江苏扬州人,实验师,博士研究生,主要研究方向:数理统计、神经网络、点云处理;颜超(1983-),男,江苏南京人,副教授,博士研究生,主要研究方向:数理统计;高琳明(1978-),女,江苏南京人,讲师,硕士,主要研究方向:图形图像学。
基金资助:
国家重点研发计划项目（2016YFD0600101）；国家自然科学基金资助项目（31770591）。

Lake extraction algorithm based on three-dimensional convolutional neural network

XU Shanshan¹, YAN Chao², GAO Linming¹

1. College of Information Science and Technology, Nanjing Forestry University, Nanjing Jiangsu 210037, China;
2. College of Meteorology and Oceanography, National University of Defense Technology, Nanjing Jiangsu 211101, China

Received:2019-04-29 Revised:2019-08-15 Online:2019-12-10 Published:2019-09-04
Contact: 徐姗姗
Supported by:
This work is partially supported by the National Key R&D Program of China (2016YFD0600101), the National Natural Science Foundation of China (31770591).

摘要/Abstract

摘要： 针对现有分析湖泊几何信息算法的二维图像湖泊轮廓提取精度低的问题，提出了一种基于三维卷积神经网络的湖泊提取算法。首先，基于平整度信息从激光扫描点云中定位出候选湖泊并对输入的候选区域点云进行体素化组织，作为神经网络的输入；同时，通过深度学习技术，从候选区域中过滤非湖泊区域；然后，基于方向链码算法从点云中提取湖泊的边缘并分析其几何形状信息。实验结果表明，所提算法在提取激光扫描点云中的湖泊精度可达到96.34%，与当前在二维图像中的湖泊提取算法相比，可对目标湖泊形状信息进行计算与分析，从而为湖泊监测与管理提供方便。

关键词: 激光扫描数据, 三维卷积神经网络, 湖泊提取, 链码, 边界描述

Abstract: Aiming at the low accuracy of lake contour extraction from two-dimensional images of the existing algorithms for analyzing the geometric information of lakes, a lake extraction algorithm based on three-dimensional convolutional neural network was proposed. Firstly, based on the flatness information, the candidate lakes were located from the laser scanning point clouds, and the candidate points were organized as voxels to be an input of the neural network. Meanwhile, the non-lake areas were filtered from candidate areas by the deep learning technique. Then, based on the chain-code algorithm, contours of lakes were extracted from point clouds and their geometry information was analyzed. The experimental results show that, the accuracy of the proposed algorithm in extracting lakes from laser scanning point clouds is 96.34%, and compared with the existing extraction algorithm for two-dimensional images, the proposed algorithm can calculate and analyze the shape information of lakes, which provides convenience for lake monitoring and management.

Key words: laser scanning data, three-dimensional convolutional neural network, lake extraction, chain-code, contour description

中图分类号:

徐姗姗, 颜超, 高琳明. 基于三维卷积神经网络的湖泊提取算法[J]. 计算机应用, 2019, 39(12): 3450-3455.

XU Shanshan, YAN Chao, GAO Linming. Lake extraction algorithm based on three-dimensional convolutional neural network[J]. Journal of Computer Applications, 2019, 39(12): 3450-3455.

参考文献

[1] 李景刚,李纪人,黄诗峰,等.Terra/MODIS时间序列数据在湖泊水域面积动态监测中的应用研究——以洞庭湖地区为例[J].自然资源学报,2009,24(5):923-933.(LI J G, LI J R, HUANG S F, et al. Application of Terra/MODIS time series data in dynamic monitoring of lake water area variations-a case study in Dongting Lake region, China[J]. Journal of Natural Resources, 2009, 24(5):923-933.)
[2] 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.(YIN Y Q, LI J G, YU T, et al. The study of object-oriented water body extraction method based on high resolution RS image[J]. Bulletin of Surveying and Mapping, 2015(1):81-85.)
[3] 骆剑承,盛永伟,沈占锋,等.分步迭代的多光谱遥感水体信息高精度自动提取[J].遥感学报,2009,13(4):604-615.(LUO J C, SHENG Y W, SHEN Z F, et al. Automatic and high-precise extraction for water information from multispectral images with the step-by-step iterative transformation mechanism[J]. Journal of Remote Sensing, 2009, 13(4):604-615.)
[4] 王碧晴,王珂,廖伟逸.遥感图像分割下的青藏高原湖泊提取[J].遥感信息,2018,33(1):117-122.(WANG B Q, WANG K, LIAO W Y. Extraction of Qinghai-Tibet plateau lake based on remote sensing image segmentation[J]. Remote Sensing Information, 2018, 33(1):117-122)
[5] 朱小强,丁建丽,夏楠,等.一种稳定阈值的湖泊水体信息提取方法[J].资源科学,2019,41(4):790-802.(ZHU X Q, DING J L, XIA N, et al. Temperature vegetation water index:a novel stabilized threshold method for lake surface water mapping[J]. Resources Science, 2019, 41(4):790-802.)
[6] 夏少波,王成,习晓环,等.ICESat-2机载试验点云滤波及植被高度反演[J].遥感学报,2014,18(6):1199-1207.(XIA S B, WANG C, XI X H, et al. Point cloud filtering and tree height estimation using airborne experiment data of ICESat-2[J]. Journal of Remote Sensing, 2014,18(6):1199-1207.)
[7] ZHOU Y, TUZEL O. Voxelnet:end-to-end learning for point cloud based 3D object detection[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:4490-4499.
[8] UY M A, LEE G H. PointNetVLAD:deep point cloud based retrieval for large-scale place recognition[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:4470-4479.
[9] QI C R, SU H, MO K, et al. PointNet:deep learning on point sets for 3D classification and segmentation[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:77-85.
[10] QI C R, YI L, SU H, et al. PointNet++:deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 2017 International Conference on Neural Information Processing Systems. New York:Curran Associates Inc., 2017:5099-5108
[11] WANG W, YU R, HUANG Q, et al. SGPN:similarity group proposal network for 3D point cloud instance segmentation[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:2569-2578.
[12] SHEN Y, FENG C, YANG Y, et al. Mining point cloud local structures by kernel correlation and graph pooling[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:4548-4557.
[13] LANDRIEU L, SIMONOVSKY M. Large-scale point cloud semantic segmentation with superpoint graphs[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:4558-4567.
[14] HUANG Q, WANG W, NEUMANN U. Recurrent slice networks for 3D segmentation of point clouds[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:2626-2635.
[15] HUANG J, YOU S. Point cloud labeling using 3D convolutional neural network[C]//Proceedings of the 3rd International Conference on Pattern Recognition. Piscataway:IEEE, 2016:2670-2675.
[16] 袁庆,楼立志,陈玮娴.基于加权总体最小二乘的平面点云拟合方法[J].测绘通报,2011(3):1-3.(YUAN Q, LOU L Z, CHEN W X. Applying weighted total least-squares to the plane point cloud fitting of terrestrial laser scanning[J]. Bulletin of Surveying and Mapping, 2011(3):1-3.)
[17] SCHNABEL R, WAHL R, KLEIN R. Efficient RANSAC for point-cloud shape detection[J]. Computer Graphics Forum, 2007, 26(2):214-226.
[18] WHITESON S, STONE P. Evolutionary function approximation for reinforcement learning[J]. Journal of Machine Learning Research, 2006, 7:877-917
[19] NAIR V, HINTON G E. Rectified linear units improve restricted Boltzmann machines[C]//Proceedings of the 27th International Conference on Machine Learning. Madison:Omnipress, 2010:807-814.
[20] KASS M, WITKIN A, TERZOPOULOS D. Snakes:active contour models[J]. International Journal of Computer Vision, 1988, 1(4):321-331.
[21] OSHER S, FEDKIW R P. Level set methods:an overview and some recent results[J]. Journal of Computational Physics, 2001, 169(2):463-502.
[22] FREEMAN H. On the encoding of arbitrary geometric configurations[J]. IRE Transactions on Electronic Computers, 1961, EC-10(2):260-268.
[23] 陆应骐,童韬.链码和在边界形状分析中的应用[J].中国图象图形学报,2002,7(12):1323-1328.(LU Y Q, TONG T. The application of chain code sum in the edge form analysis[J]. Journal of Image and Graphics, 2002, 7(12):1323-1328.)
[24] 梁欣廉,张继贤,李海涛,等.激光雷达数据特点[J].遥感信息,2005(3):71-76.(LIANG X L, ZHANG J X, LI H T, et al. The characteristics of LIDAR data[J]. Remote Sensing Information, 2005(3):71-76.)
[25] VEDALDI A, KAREL L. MatConvNet:convolutional neural networks for Matlab[C]//Proceedings of the 23rd ACM International Conference on Multimedia. New York:ACM, 2015:689-692.

基于三维卷积神经网络的湖泊提取算法

Lake extraction algorithm based on three-dimensional convolutional neural network

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

[1]	刘宏宇, 梁秀波, 吴俊涵. 基于Kubernetes的Fabric链码管理及高可用技术[J]. 计算机应用, 2021, 41(4): 956-962.
[2]	徐少伟, 秦品乐, 曾建朝, 赵致楷, 高媛. 基于注意力机制的两阶段纵膈淋巴结自动分割算法[J]. 计算机应用, 2021, 41(2): 556-562.
[3]	陈莉, 王洪元, 张云鹏, 曹亮, 殷雨昌. 联合均等采样随机擦除和全局时间特征池化的视频行人重识别方法[J]. 计算机应用, 2021, 41(1): 164-169.
[4]	万永菁, 王博玮, 娄定风. 基于三维卷积神经网络的虫音特征识别方法[J]. 计算机应用, 2019, 39(9): 2744-2748.
[5]	金忠星, 李东. 消费者偏好预测的深度学习神经网络模型[J]. 计算机应用, 2019, 39(7): 1888-1893.
[6]	魏巍, 段晓东, 刘勇奎, 郭晨. 一种新压缩顶点链码[J]. 计算机应用, 2017, 37(6): 1747-1752.
[7]	徐义钊白瑞林余震虹吉峰. 基于链码跟踪的Data Matrix二维条码快速识别[J]. 计算机应用, 2012, 32(11): 3206-3209.
[8]	吴少泓王云宽孙涛李兵. 基于距离分布直方图的数字识别算法[J]. 计算机应用, 2012, 32(08): 2299-2304.
[9]	刘巍金文标肖仙谦. 基于主元分析和聚类的直线检测算法[J]. 计算机应用, 2011, 31(05): 1202-1204.
[10]	杨闯陈家新黎蔚. 基于中值夹角链码的掌静脉特征提取[J]. 计算机应用, 2009, 29(11): 3048-3050.
[11]	王萍强兆庆许晋玮贾惠珍. 基于链码描述的图像图形特征提取[J]. 计算机应用, 2009, 29(08): 2065-2067.
[12]	刘韶军王库. 基于自适应链码的棉种周长测量[J]. 计算机应用, 2008, 28(11): 2881-2883.
[13]	尚振宏付晓东袁梅宇刘辉. 基于欧氏距离的实时直线检测算法[J]. 计算机应用, 2008, 28(1): 177-180.
[14]	王萍芮乾梅贾惠珍. 基于链码信息的雹云识别及冰雹落区定位[J]. 计算机应用, 2007, 27(2): 446-448.
[15]	沈叶健徐守时 . 一种有效的可见光图像中水坝目标的识别方法[J]. 计算机应用, 2006, 26(8): 1972-1974.