基于无人机影像的城市植被精细分类

摘要
图/表
参考文献
相关文章 (6)

全文: PDF (1625 KB) HTML (0 KB)
输出: BibTeX | EndNote (RIS)

摘要针对传统植被资源调查方法工作量大、成本高、效率低的问题,利用高分辨率无人机遥感影像,联合地物光谱-纹理-空间信息,构建了一种适用于描述城市不同植被种类的多维特征空间,在此基础上对三种应用广泛的分类算法(基于像素的、面向对象的支持向量机及深度学习Mobile-Unet语义分割模型)开展了对比分析研究.结果表明:本文提出的联合地物光谱-纹理-空间信息的特征空间构建方法能够有效地描述城市不同类型植被的特征差异,提升影像分割、植被分类的精度;在分类精度上,基于像素和面向对象的支持向量机分类结果的总体精度均超过90%,深度学习方法的总体分类精度为84%;在算法效率上,传统机器学习方法也优于深度学习方法.因此,得出结论针对城市小区域、小样本的植被精细分类,传统机器学习分类方法比深度学习方法效果更好.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	林怡
	张文豪
	宇洁
	张翰超

关键词 ：无人机影像, 城市植被, 精细分类, 多维特征空间, 机器学习, 深度学习

Abstract：The fine classification and information extraction of urban public vegetation based on remote sensing methods hasgreat significance in urban land use change, urban ecological environment monitoring and urban planning. Aiming toreducethe heavy workload, high cost and enhancetheefficiency of traditional vegetation resource survey methods, a multi-dimensional feature space including spectral, spatial and texture features for describing different types of urban vegetation was proposed in this study. Then a comparison experiment of three typical classification algorithms (pixel-based, object-oriented support vector machine and deep learning semantic segmentation model-Mobile-Unet) was accomplished. The experimental results indicate that:The proposed spectrum-texture-spatial feature extraction method can describe different types of urban vegetation effectively, and improve the accuracy of image segmentation and vegetation classification significantly. In terms of classification accuracy, the overall accuracy of pixel-based and object-oriented support vector machine classification exceed 90%, while the overall accuracy of deep learning method is only 84%. In terms of efficiency, traditional machine learning methods are also superior to deep learning method. Therefore, for the fine classification of different vegetation in small urban areas with small samples, traditional machine learning classification methods are more effective than deep learning method.

Key words： unmanned aerial vehicle image urban vegetation fine classification multi-dimensional feature space machine learning deep learning

收稿日期: 2021-11-17

PACS:	TP751.1
	X87

基金资助:地理信息工程国家重点实验室开放基金课题(SKLGIE2018- K-3-1);上海市"科技创新行动计划"科技攻关项目(20dz1201200);自然资源部测绘科学与地球空间信息技术重点实验室开放研究基金课题(2020-2-3)

通讯作者: 宇洁,助理教授,2011_jieyu@tongji.edu.cn E-mail: 2011_jieyu@tongji.edu.cn

作者简介: 林怡(1970-),女,浙江温州人,研究员,博士,主要从事数字摄影测量与城市、环境遥感研究.发表论文80余篇.

引用本文:

林怡, 张文豪, 宇洁, 张翰超. 基于无人机影像的城市植被精细分类[J]. 中国环境科学, 2022, 42(6): 2852-2861. LIN Yi, ZHANG Wen-hao, YU Jie, ZHANG Han-chao. Fine classification of urban vegetation based on UAV images. CHINA ENVIRONMENTAL SCIENCECE, 2022, 42(6): 2852-2861.

链接本文:

http://www.zghjkx.com.cn/CN/ 或 http://www.zghjkx.com.cn/CN/Y2022/V42/I6/2852

[1]	CJJ/T 85-2017城市绿地分类标准[S].CJJ/T 85-2017 Standard for classification of urban green space[S].
[2]	白晓琼,王汶,林子彦,等.基于高空间分辨率遥感影像的三维绿度度量[J].国土资源遥感, 2019,31(4):53-59.Bai X Q, Wang W, Lin Z Y, et al. Three-dimensional measuring for green space based on high spatial resolution remote sensing images[J]. Remote Sensing for Land and Resources, 2019,31(4):53-59.
[3]	费鲜芸.高分辨率遥感影像在城市绿地信息提取中的应用研究[D].泰安:山东农业大学, 2006.Fei X Y. The study of urban green space surveying based on high resolution RS images[D]. Taian:Shandong Agricultural University, 2006.
[4]	赵康兵.高分辨率遥感影像技术应用研究——以城市绿化精细化调查为例[J].南京工程学院学报(自然科学版), 2019,17(4):50-56.Zhao K B. Application of high resolutionremote sensing image technology-taking urban greening investigation as an example[J]. Journal of Nanjing Institute of Technology (Natural Science Edition), 2019,17(4):50-56.
[5]	颜梅春.高分辨率影像的植被分类方法对比研究[J].遥感学报, 2007,(2):235-240.Yan M C. Research and contrast on several vegetation-classification methods of high-resolution satellite image data[J]. Journal of Remote Sensing, 2007,(2):235-240.
[6]	韩文霆,张立元,张海鑫,等.基于无人机遥感与面向对象法的田间渠系分布信息提取[J].农业机械学报, 2017,48(3):205-214.Han W T, Zhang L Y, Zhang H X, et al. Extraction method of sublateral canal distribution information based on UAV remote sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017,48(3):205-214.
[7]	Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(4):640-651.
[8]	Badrinarayanan V, Kendall A, Cipolla R.SegNet:A deep convoluntional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(12):2481-2495.
[9]	Ronneberger O, Fischer P, Brox T.U-Net:Convolutional networks for biomedicalimage segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2015:234-241.
[10]	Li D, Ke Y, GongH, et al. Object-based urban tree species classification using bi-temporal WorldView-2 and World View-3images[J]. Remote Sensing, 2015,7(12):16917-16937.
[11]	徐知宇,周艺,王世新,等.面向GF-2遥感影像的U-Net城市绿地分类[J].中国图象图形学报, 2021,26(3):0700-0713.Xu Z Y, Zhou Y, Wang S X, et al. U-Net for urban green space classification in Gaofen-2remote sensingimages[J]. Journal of Image and Graphics, 2021,26(3):0700-0713.
[12]	周涛,胡振琪,韩佳政,等.基于无人机可见光影像的绿色植被提取[J].中国环境科学, 2021,41(5):2380-2390.Zhou T, Hu Z Q, Han J Z, et al. Green vegetation extraction based on visible light image of UAV[J]. China Environmental Science, 2021, 41(5):2380-2390.
[13]	杨蜀秦,宋志双,尹瀚平,等.基于深度语义分割的无人机多光谱遥感作物分类方法[J].农业机械学报, 2021,52(3):185-192.Yang S Q, Song Z S, Yin H P, et al. Crop classification method of UVA multispectral remote sensing based on deep semantic segmentation[J]. Transactions of the Chinese Society for Agricultural Machinery, 2021,52(3):185-192.
[14]	Torres D L, Feitosa R Q, Happ P N, et al. Applying fully convolutional architectures for semantic segmentation of a single tree species in urban environment on high resolution UAV optical imagery[J]. Sensors, 2020,20(2):563.
[15]	戴鹏钦,丁丽霞,刘丽娟,等.基于FCN的无人机可见光影像树种分类[J].激光与光电子学进展, 2020,57(10):36-45.Dai P Q, Ding L X, Liu L J, et al. Tree species identification based on FCN using the visible images obtained from an unmanned aerial vehicle[J]. Laser& Optoelectronics Progress, 2020,57(10):36-45.
[16]	林志玮,涂伟豪,黄嘉航,等.基于FC-DenseNet的低空航拍光学图像树种识别[J].国土资源遥感, 2019,31(3):225-233.Lin Z W, Tu W H, Huang J H, et al. Tree species recognition of UAV aerial images based on FC-DenseNet[J]. Remote Sensing for Land and Resources, 2019,31(3):225-233.
[17]	Schiefer F, Kattenborn T, Frick A, et al.U-Net:Mapping forest tree species in high resolution UAV-based RGB-imagery by means of convolutional neural networks[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020,170:205-215.
[18]	Sothe C, De Almeida C M, Schimalski M B, et al.Comparative performance of convolutional neural network, weighted and conventional support vector machine and random forest for classifying tree species using hyperspectral and photogrammetric data[J]. GIS& Remote Sensing, 2020,57(3):369-394.
[19]	左萍萍,付波霖,蓝斐芜,等.基于无人机多光谱的沼泽植被识别方法[J].中国环境科学, 2021,41(5):2399-2410.Zuo P P, Fu B L, Lan F W, et al. Classification method of swamp vegetation using UAV multispectral data[J]. China Environmental Science, 2021,41(5):2399-2410.
[20]	李艳,和太平.广东省茂名市居住区绿地植物资源及应用调查分析[J].安徽农业科学, 2021,49(8):116-119.Li Y, He T P.Investigation and application analysis of green land plantresources in residential area of Maoming city, Guangdong Province[J]. Journal of Anhui Agricultural Sciences, 2021,49(8):116-119.
[21]	杨礼.融合UAV遥感影像与SFM点云的树木识别及参数提取[D].焦作:河南理工大学, 2018.Yang L. Tree parameters extraction using UAV remote sensing image and SFM point cloud[D]. Jiaozuo:Henan Polytechnic University, 2018.
[22]	汪小钦,王苗苗,王绍强,等.基于可见光波段无人机遥感的植被信息提取[J].农业工程学报, 2015,31(5):152-159.Wang X Q, Wang M M, Wang S Q, et al. Extraction of vegetation information from visible unmanned aerial vehicle images[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015,31(5):152-159.
[23]	Verrelst J, Schaepman M E, Koetz B, et al.Angular sensitivity analysis of vegetation indices derived from CHRIS/PROBA data[J]. Remote Sensing of Environment, 2008,112(5):2341-2353.
[24]	Meyer G E, Neto J C. Verification of color vegetation indices for automated crop imaging applications[J]. Computers& Electronics in Agriculture, 2008,63(2):282-293.
[25]	宋荣杰,宁纪锋,刘秀英,等.基于纹理特征和SVM的QuickBird影像苹果园提取[J].农业机械学报, 2017,48(3):188-197.Song R J, Ning J F, Liu X Y, et al. Apple orchard extraction with Quick Birdimagery based ontexture features and support vector machine[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017,48(3):188-197.
[26]	郑亚卿,杨张海,高凯旋,等.基于多时相Landsat-8的森林生态系统树种精细分类[J].测绘与空间地理信息, 2021,44(6):158-161.Zheng Y Q, Yang Z H, Gao K X, et al. Sophisticated classification of tree species in forest ecosystembased on multi-temporal Landsat-8[J]. Geomatics& Spatial Information Technology, 2021,44(6):158-161.
[27]	Vapnik V N. The nature of statistical learning theory[M]. New York:Springer-verlag, 1995.
[28]	边肇祺,张学工.模式识别第二版[M].北京:清华大学出版社, 2000.Bian Z Q, Zhang X G. Pattern recognition (The second edition)[M]. Beijing:Tsinghua University Press, 2000.
[29]	王志盼,沈彦,王亮,等.单类分类框架下的高分辨率遥感影像建筑物变化检测算法[J].武汉大学学报(信息科学版), 2020,45(10):1610-1618.Wang Z P, Shen Y, Wang L, et al. High resolutionremote sensing image building change detection basedon one class classifier framework[J]. Geomatics and Information Science of Wuhan University, 2020,45(10):1610-1618.
[30]	Baatz M, Schape A. Object-oriented and multi-scale image analysis in semantic networks[C]//Proceedings of the 2nd International Symposium on Operationalization of Remote Sensing, Enschede, Netherlands, 1999:16-20.
[31]	周小成,郑磊,黄洪宇.基于多特征优选的无人机可见光遥感林分类型分类[J].林业科学, 2021,57(6):24-36.Zhou X C, Zheng L, Huang H Y. Classification of forest stand based on multi-feature optimization of UAV visible light remote sensing[J]. Scientia Silvae Sinicae, 2021,57(6):24-36.
[32]	Howard A G, Zhu M, Chen B, et al. MobileNets:Efficient convolutional neural networks for mobile vision applications[J]. arXiv:1704.04861, 2017.