Application of Hyperspectral Data to the Classification and Identification of Severity of Wheat Stripe Rust
WANG Hai-guang1, MA Zhan-hong1, WANG Tao2, CAI Cheng-jing1, AN Hu1, ZHANG Lu-da2*
1. Department of Plant Pathology, China Agricultural University, Beijing 100094,China 2. College of Science, China Agricultural University, Beijing 100094,China
摘要： 小麦条锈病是我国小麦生产上造成损失最大、危及范围最广的一种病害，对该病的监测预报是实施有效治理措施的重要基础和依据。文章以88个小麦叶片为试验材料，其中条锈病叶按严重度分为8级，健康小麦叶片为对照，由ASD Field-Spec Pro FR 2500型光谱仪和LI-Cor1800-12外置积分球获取高光谱数据，采用SVM算法对不同严重度的小麦条锈病病叶进行了判别分析。按1∶1比例随机划分样品集，校正集的44个样品建立模型，对预测集的44个样品的严重度进行预测识别，总体正确识别率达97%，表明SVM算法用于小麦条锈病严重度分级识别是可行的。
关键词：高光谱;小麦条锈病;分级识别;支持向量机
Abstract：Wheat stripe rust, caused by Puccinia striiformis f.sp.tritici, is one of pandemic diseases causing severe losses in China. Monitoring and warning of this disease is principal for its precise prediction and for implementing effective measures to control it. The hyperspectral data used for analysis were attained from 88 leaves including healthy leaves and infected leaves over a range of disease severity levels. Support vector machine (SVM) was applied to classify and identify the severity of wheat leaves infected by the pathogen. The model was built based on 44 proof-read samples to estimate 44 proof-test samples. And the identification accuracy is totally 97%. So SVM can be used in the classification and identification of severity of wheat stripe rust based on attained hyperspectral data.
Key words：Hyperspectra;Wheat stripe rust;Classification and identification;Support vector machine
王海光1,马占鸿1,王 韬2,蔡成静1,安虎1,张录达2* . 高光谱在小麦条锈病严重度分级识别中的应用[J]. 光谱学与光谱分析, 2007, 27(09): 1811-1814.
WANG Hai-guang1, MA Zhan-hong1, WANG Tao2, CAI Cheng-jing1, AN Hu1, ZHANG Lu-da2* . Application of Hyperspectral Data to the Classification and Identification of Severity of Wheat Stripe Rust. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2007, 27(09): 1811-1814.
[1] Malthus T J, Maderia A C. Remote Sensing of Environment, 1993, 45: 107. [2] Adams M L, Philpot W D, Norvell W A, et al. International Journal of Remote Sensing, 1999, 20(18): 3663. [3] Steddom K, Heidel G, Jones D, et al. Phytopathology, 2003, 93: 720. [4] Nutter F W, Jr, Guan J, Gotlieb A R. et al. Plant Disease, 2002, 86: 269. [5] Guan J, Nutter F W, Jr. Canadian Journal of Plant Pathology, 2003, 25: 143. [6] WU Shu-wen, WANG Ren-chao, CHEN Xiao-bin, et al(吴曙雯, 王人潮, 陈晓斌, 等). Journal of Shanghai Jiaotong University(Agricultural Science) (上海交通大学学报·农业科学版),2002,20(1):73, 84. [7] Nutter F W, Jr, Tylka G L, Guan J, et al. Journal of Nematology, 2002, 34(3): 222. [8] HUANG Mu-yi, HUANG Yi-de, HUANG Wen-jiang, et al(黄木易, 黄义德, 黄文江, 等). Journal of Anhui Agricultural Sciences(安徽农业科学), 2004, 32(1): 132. [9] ZHANG Hong-ming(张宏名). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，1994，14(5)：25. [10] Vapnik V N. The Nature of Statistical Learning Theory. New York: Springer-Verlag, 1995. [11] Cortes C, Vapnik V. Machine Learning, 1995, 20(3): 273. [12] ZHANG Lu-da, JIN Ze-chen, SHEN Xiao-nan, et al(张录达，金泽宸，沈晓南，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2005，25(9)：1400. [13] WANG Liang-shen, ZHU Yu-cai, CHEN Shao-hua, et al(王亮申，朱玉才，陈少华，等). Computer Engineering and Design(计算机工程与设计)，2005，26(9)：2453. [14] LIU Shu-hua, ZHANG Xue-gong, ZHOU Qun, et al(刘沐华，张学工，周 群，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2006, 26(4): 629. [15] MENG Geng-xiang, FANG Jing-long(蒙庚祥，方景龙). Computer Engineering and Design(计算机工程与设计)，2005，26(6)：1592，1598. [16] WANG Huan-liang, HAN Ji-qing, ZHANG Lei(王欢良，韩纪庆，张 磊). Journal of Harbin Institute of Technology(哈尔滨工业大学学报)，2003，35(4)：389. [17] ZHANG Lu-da, SU Shi-guang, WANG Lai-sheng, et al(张录达, 苏时光, 王来生, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2005, 25(1): 33. [18] LUO Jian-cheng, ZHOU Cheng-hu, LEUNG Yee, et al(骆剑承, 周成虎, 梁 怡, 等). Journal of Remote Sensing(遥感学报), 2002, 6(1): 50.
