FANG Xiufeng,TAN Bingxiang*,DU Huaqiang,et al.Suitability of band width for Hyperion image classification based on wavelet analysis[J].Journal of Nanjing Forestry University(Natural Science Edition),2018,42(05):141-147.[doi:10.3969/j.issn.1000-2006.201710042]





Suitability of band width for Hyperion image classification based on wavelet analysis
1.中国林业科学研究院资源信息研究所,北京 100091; 2.浙江农林大学环境与资源学院,浙江 杭州 311300; 3.吉林省白河林业局,吉林 延边 133613
FANG Xiufeng1 TAN Bingxiang1* DU Huaqiang2 WANG Huaijing1 LI Taixing3
1. Institution of Forest Resources Information Technique, Chinese Academy of Forestry, Beijing 100091, China; 2. School of Enviromental and Resources Science, Zhejiang A & F University, Hangzhou 311300, China; 3. Baihe Forestry Bureau of Jilin Provinc
Hyperion影像 小波分解 小波融合 林分分类识别
Hyperion image wavelet decomposition wavelet fusion classification and identification of forest
S758; TP79
【目的】高光谱遥感为地物的精细识别提供优势的同时,也带来了数据量多、波段间相关性大、处理精度和效率下降等问题,而且在遥感分类中并不是使用的通道越多、波段越窄效果越好。因此笔者从光谱角度出发探讨降低高光谱数据量,以寻求适宜遥感分类波段宽度的方法。【方法】首先对Hyperion影像进行处理,主要包括去除未定标和受水汽影像波段、坏线、条纹和Smile效应,辐射定标和大气校正处理后得到161个波段,对选用的LIR级数据进行几何校正。根据样地调查情况确定试验区待分类别,对提取的14类地物样本平均光谱进行7次Sym3小波分解,由得到的小波细节系数方差和小波细节系数熵分析适宜各类型识别的光谱区间,然后将不同光谱区间内窄波段进行小波融合,最后选取支持向量机方法进行分类识别。【结果】美人松林、落叶松林、樟子松林、针叶混交林、阔叶混交林、火烧迹地、水体、耕地和未利用地9类地物识别的适宜光谱分辨率为40 nm,剩余5种地物识别的适宜光谱分辨率为80 nm,不同光谱区间对应的波段数大大降低,且最终分类精度总体都达到81%以上。【结论】将小波分析与支持向量机方法(SVM)结合,解决了高光谱存在的“维数灾难”问题,提高了高光谱数据的利用率,遥感分类中并不是使用的通道越多、波段越窄效果越好。
【Objective】Hyperspectral remote sensing is advantageous for fine identification of features, but has the problem of large data volume, high correlations among bands, and low processing accuracy and efficiency. In remote sensing classification, it is not the case that the use of a greater number of narrow bands ensures better classification perfor-mance. Therefore, the primary objective of the current work was to determine an appropriate bandwidth of Hyperion images by exploring the reduction of hyperspectral data, from the perspective of spectra.【Method】In this study, the preprocessing procedures for Hyperion images principally included the removal of uncalibrated and water vapor affected bands, eliminating the bad lines, stripes and smile effects, radiometric calibration and atmospheric correction. After this processing, 161 bands were retained to implement further geometric correction. Fourteen land use classes or schemes in the experimental area were selected according to field sample plot inventories, and a 7-scale Sym 3 wavelet decomposition was carried out on the average spectra of the 14 land use types. The spectral bands suitable for each type of recognition were mainly concentrated around 40 and 80 nm because of the variance and entropy of the wavelet detail coefficients. The narrow bands identified in the different spectral intervals were fused using the wavelet function to classify, by implementation, the support vector machines(SVM)algorithm.【Result】It was concluded that the appropriate spectral resolution to identify nine types of ground objects, such as Pinus syluestriformis forest, Larix gmelinii forest, Pinus sylvestris forest, mixed conifers forest, mixed broadleaf forest, fire burned area, water, arable land, and unused land, is 40 nm and the suitable spectral resolution for the remaining five kinds of terrain recognitions is 80 nm. The number of bands in different spectral intervals was greatly reduced. Further, the final classification accuracy was over 81%. 【Conclusion】The combination of wavelet analysis and SVM was helpful to solve the “dimensional disaster” issue with hyperspectral data, which indicates that the proposed method is feasible.


[1] 杜培军, 夏俊士, 薛朝辉,等. 高光谱遥感影像分类研究进展[J]. 遥感学报, 2016, 20(2):236-256. DOI: 10.11834/jrs.20165022.
DU P J, XIA J S, XUE Z H, et al. Review of hyperspectral remote sensing image classification[J]. Journal of Remote Sensing, 2016, 20(2):236-256.
[2] 李显彬, 姜小光, 刘亮,等. 基于光谱重建的高光谱特征参数选择方法——以苏北地区Hyperion数据为例[J]. 遥感学报, 2007, 11(4):589-594.DOI: 10.3321/ j.issn:1007-4619.2007.04.022.
LI X B, JIANG X G, LIU L, et al. A new feature selection method for EO-1 Hyperion image classification: a case study of Subei region, China[J]. Journal of Remote Sensing, 2007, 11(4):589-594.
[3] 林志垒, 晏路明. 高光谱影像的M-ICA地物识别算法与应用[J]. 地球信息科学学报, 2011, 13(1):126-132.DOI:10.3724/SP.J.1047.2011.00126.
LIN Z L, YAN L M. Object recognition algorithm and its application on hyperspectral imagery based on M-ICA[J]. Geo-information Science,2011,13(1):126-132.
[4] LUCAS R, BUNTING P, PATERSON M, et al. Classification of Australian forest communities using aerial photography, CASI and HyMap data[J]. Remote Sensing of Environment, 2008, 112(5):2088-2103.DOI:10.1016/j.rse.2007.10.011.
[5] WARNER T A, SHANK M C. Spatial autocorrelation analysis of hyperspectral imagery for feature selection [J]. Remote Sensing of Environment, 1997, 60(1):58-70.DOI:10.1016/S0034-4257(96)00138-1.
[6] 高恒振. 高光谱遥感图像分类技术研究[D]. 北京:国防科学技术大学,2011.
GAO H Z. Rasearch on classification techinique for hyperspectral remote sensing imagery[D]. Beijing: National University of Defense Technology,2011.
[7] 柳萍萍, 林辉, 孙华,等. 高光谱数据的降维处理方法研究[J].中南林业科技大学学报,2011, 31(11):34-38. DOI: 10.3969/j.issn.1673-923X.2011.11.007.
LIU P P, LIN H, SUN H, et al. Dimensionality reduction method of hyperion EO-1 data[J]. Journal of Central South University of Forestry and Technology, 2011, 31(11):34-38.
[8] 陈刚, 陈小梅, 李婷,等. 基于小波分解的光谱特征提取算法研究[J]. 光谱学与光谱分析, 2010, 30(11):3027-3030. DOI:10.3964/j.issn.1000-0593(2010)11-3027-04.
CHEN G, CHEM X M, LI T, et al. Research on spectral data feature extraction based on wavelet decomposition[J]. Spectroscopy and Spectral Analysis, 2010, 30(11):3027-3030.
[9] KAEWPIJIT S, MOIGNE J L, El-GHAZAWI T. Automatic reduction of hyperspectral imagery using wavelet spectral analysis[J]. IEEE Transactions on Geoscience & Remote Sensing, 2003, 41(4):863-871. DOI:10.1109/TGRS.2003.810712.
[10] 刘燕德, 欧阳爱国, 应义斌. 小波分析用于光谱信号处理及其在Matlab中的实现[J]. 传感技术学报, 2006, 19(3):821-823. DOI: 10.3969/j.issn.1004-1699. 2006.03.071.
LIU Y D, OUYANG A G, YING Y B. Application of wavelet analysis in signal process using matlab[J]. Chinese Journal of Senior and Actuators, 2006, 19(3):821-823.
[11] GONZALEZ-AUDICANA M, SALETA J L, CATALAN R G, et al. Fusion of multispectral and panchromatic images using improved IHS and PCA mergers based on wavelet decomposition[J]. Geoscience & Remote Sensing IEEE Transactions on, 2004, 42(6):1291-1299.DOI:10.1109/TGRS.2004.825593.
[12] 谭炳香, 李增元, 陈尔学, 等. EO-1 Hyperion 高光谱数据的预处理[J].遥感信息, 2005(6):36-41. DOI:10.3969/j.issn.1000-3177.2005.06.010.
TAN B X, LI Z Y, CHEN E X, et al. Preprocessing of EO-1 Hyperion hyperspectral data[J]. Remote Sensing Information, 2005(6):36-41.
[13] 梁继, 王建. Hyperion高光谱影像的分析与处理[J]. 冰川冻土, 2009, 31(2):63-69.
LIANG J, WANG J. Hyperion hyperspectral image: analysis and process[J]. Journal of Glaciology and Geocryology, 2009, 31(2):63-69.
[14] ZHU R H, WAN M, FAN G B. An image fuse method based on pyramid transformation[J]. Computer Simulation, 2007, 24(12):178-181.
[15] 郭志强. 基于区域特征的小波变换图像融合方法[J]. 武汉理工大学学报, 2005, 27(2):65-67. DOI: 10.3321/j.issn:1671-4431.2005.02.020.
GUO Z Q. Wavelettransform image fusion based on regional features[J].Journal of Wuhan University of Technology, 2005, 27(2):65-67.
[16] 桑琦, 陈浩, 李卫华. 基于小波变换的图像融合方法仿真[J]. 电子科技, 2017, 30(3):1-3. DOI: 10.16180/j.cnki.issn1007-7820.2017.03.001.
SANG Q, CHEN H, LI W H. Image fusion method simulation based on wavelet transform[J]. Electronic Science and Technology, 2017, 30(3):1-3.
[17] ROSSO O A, BLANCO S, YORDANOVA J, et al. Wavelet entropy: a new tool for analysis of short duration brain electrical signals[J]. Journal of Neuroscience Methods, 2001, 105(1):65-75. DOI: 10.1016/S0165-0270(00)00356-3.
[18] 陈玉明, 吴克寿, 李向军. 一种基于信息熵的异常数据挖掘算法[J]. 控制与决策, 2013(6):867-872.
CHEN Y M, WU K S, LI X J. A kind of outlier mining algorithm based on information entropy[J]. Control and Decision, 2013(6):867-872.
[19] 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.


收稿日期:2017-10-27 修回日期:2018-06-24 基金项目:浙江省省院合作林业科技项目(2017SY04); 中国林科院基本科研业务费专项(CAFYBB2017MB012) 第一作者:房秀凤(1156807749@qq.com)。*通信作者:谭炳香(tan@ifrit.ac.cn),研究员。
更新日期/Last Update: 2018-09-15