基于高光谱影像多维特征的植被精细分类

摘要/Abstract

摘要： 目前,高光谱植被精细分类存在三个问题:单纯利用光谱信息得到的分类精度较低;光谱数据存在噪声影响了最终的分类结果;
缺少针对具体应用场景而设计的分类方法。为此,提出了一种基于高光谱影像多维特征的植被精细分类方法,通过光谱
数据降维、纹理特征提取以及植被指数选择三个方面对高光谱影像数据进行分析与利用,依靠前期现场调查得到的地面
植被分布情况,选择训练样本并进行支持向量机(Support vector machine, SVM)监督分类,完成地面植被的精细分类,
对分类结果进行验证,总体精度可达99.6\%。结果表明,基于高光谱影像多维特征的植被分类方法能够有效地减小数据噪声、
提高信息利用率,为植被生态监测提供更为准确的数据支撑。

关键词: 高光谱, 光谱降维, 纹理特征, 植被指数, 支持向量机

Abstract: At present, there are three major challenges in the sophisticated vegetation classification using hyperspectral image.
The first is that the accuracy of classification obtained simply by using spectral information is low. The second
is that the presence of noise in the spectral data affects the final classification results, and the third is
the lack of classification methods designed for specific application scenarios. To this end, a method for
sophisticated vegetation classification based on multi-dimensional features of hyperspectral images is
proposed. In this method, hyperspectral image data are analyzed and utilized firstly through three
aspects of spectral data dimension reduction, texture feature extraction and vegetation index selection.
And then, based on the distribution of ground vegetation obtained from previous field surveys, training
samples are selected and Support Vector Machine (SVM) supervised classification is performed, which
results in the sophisticated classification of ground vegetation at last. To verify the classification
results, the overall accuracy can reach 99.6\%. The result shows that vegetation classification based on
multi-dimensional features of hyperspectral image can effectively reduce data noise and improve information
utilization rate, and can provide more reliable data support for vegetation ecological monitoring work.

Key words: hyperspectral, spectral dimension reduction, texture feature, vegetation index, support vector machine

苗宇宏, 杨敏, 吴国俊, . 基于高光谱影像多维特征的植被精细分类[J]. 大气与环境光学学报, 2020, 15(2): 117-124.

参考文献 0

	[1]
	Chen Dan. Vegetation Classification Based on the HIS Hyperspectral Data of HJ-IA Satellite [D].
	Nanjing: Master Thesis of Nanjing Agricultural University of China, 2012 (in Chinese).
	陈丹. 基于HJ-1A星HSI高光谱数据的植被分类研究[D]. 南京:南京农业大学硕士论文, 2012.
	[2]
	Pu Ruiliang, Gong Peng. Hyperspectral Remote Sensing and Its Application [M]. Beijing: Higher Education Press, 2000 (in Chinese).
	浦瑞良, 宫鹏. 高光谱遥感及其应用[M]. 北京:高等教育出版社, 2000.
	[3]
	Liang Zhilin, Zhang Liyan, Zeng Xianling, et al. Hyperspectral remote sensing for urban vegetation identification [J].
	Geospatial Information, 2017, 15(2): 72-75 (in Chinese).
	梁志林, 张立燕, 曾现灵, 等. 高光谱遥感城市植被识别方法研究[J]. 地理空间信息, 2017, 15(2): 72-75.
	[4]
	Ming Qunjie. Identification and Extraction of Typical Vegetation on Qinghai-Tibetan Plateau [D].
	Beijing: Master Thesis of China University of Geosciences of China, 2017 (in Chinese).
	明群杰. 基于光谱匹配技术的青藏高原典型植被识别与提取[D]. 北京:中国地质大学硕士论文,2017.
	[5]
	Sylvain J, Mireille G. A novel maximum likelihood based method for mapping depth and water quality from hyperspectral
	remote-sensing data [J]. Remote Sensing of Environment, 2014, 147(18): 121–132.
	[6]
	Cheng Boyan, Liu Qiang, Li Xiaowen, et al. Building simplification using backpropagation neural networks: a combination
	of cartographers' expertise and raster-based local perception [J]. Mapping Sciences and Remote Sensing, 2013, 50(5): 527-542.
	[7]
	Zhou Y M, Zhang R Q, Ma H Y, et al. Retrieving of salt lake mineral ions salinity from hyper-spectral data
	based on BP neural network [J]. Remote Sensing for Land and Resources, 2016, 28(2): 34-40 (in Chinese).
	周亚敏, 张荣群, 马鸿元,等. 基于BP神经网络的盐湖矿物离子含量高光谱反演[J]. 国土资源遥感, 2016, 28(2): 34-40.
	[8]
	Maulik U, Chakraborty D. Learning with transductive SVM for semisupervised pixel classification of remote sensing imagery [J].
	Isprs Journal of Photogrammetry \& Remote Sensing, 2013, 77: 66-78.
	[9]
	Liu Yanling. Physicochemical Parametric Inversion and Refined Classification of Vegetation based on Hyperspectral Remote
	Sensing Image [D]. Harbin: Master Thesis of Harbin Institute of Technology of China, 2018 (in Chinese).
	刘艳玲. 基于高光谱图象的植被理化参数反演及精细分类[D]. 哈尔滨:哈尔滨工业大学硕士论文, 2018.
	[10]
	Luo Guangchun, Chen Guangyi, Tian Ling, et al. Minimum noise fraction versus principal component analysis as a
	preprocessing step for hyperspectral imagery denoising [J]. Canadian Journal of Remote Sensing, 2016, 42(2): 106-116.
	[11]
	Wang C W, Wang H W, Hu B, et al. A new spectral-spatial algorithm method for hyperspectral image target detection [J].
	Spectroscopy and Spectral Analysis, 2016, 36(4): 1163-1169 (in Chinese).
	王彩文, 王洪伟, 胡炳樑, 等. 一种新的空谱联合探测高光谱影像目标探测算法[J]. 光谱学与光谱分析, 2016, 36(4): 1163-1169.
	[12]
	Qiao Yu. Study on Spectral Reflectance Characteristics of Representative Vegetation and Its Application in The Middle
	Section of The Qilian Mountains [D]. Lanzhou: Master Thesis of Lanzhou University of China, 2017 (in Chinese).
	乔雨. 祁连山中段典型植被的光谱特征研究与应用[D]. 兰州:兰州大学硕士论文, 2017.
	[13]
	Rouse J W, Haas R H, Schell J A, et al. Monitoring Vegetation Systems in the Great Plains with ERTS[C]//
	Proceedings of Third Earth Resources Technology Satellite-1 Symposium. Greenbelt, 1974 (351): 310-317.
	[14]
	Villamuelas M, Fernandez N, Albanell E, et al. The enhanced vegetation index (EVI) as a proxy for diet quality and
	composition in a mountain ungulate [J]. Ecological Indicators, 2016, 61: 658-666.
	[15]
	Rondeaux G, Steven M, Baret F. Optimization of soil-adjusted vegetation indices [J].
	Remote Sensing of Environment, 1996, 55(2): 95-107.
	[16]
	Elvidge C D, Chen Z K. Comparison of broad-band and narrow-band red and near-infrared vegetation indices [J].
	Remote Sensing of Environment, 1995, 54(1): 38-48.