棉花黄萎病病叶光谱特征与病情严重度的估测

Spectrum Characteristics of Cotton Single Leaf Infected by Verticillium Wilt and Estimation on Severity Level of Disease

【目的】阐明棉花黄萎病病叶光谱特征并对其病情严重度进行估测,为今后通过航空、航天遥感大面积监测棉花黄萎病提供依据。【方法】试验采用不同品种棉花黄萎病的病叶材料,在不同发病时期的病谱田和大田同步测定其光谱和发病严重度,定性和定量地分析病叶光谱反射特征、微分光谱特征差异。【结果】棉花不同品种、不同发病时期的黄萎病的病叶光谱均随发病严重度的增加而表现出有规律的变化,可见光(400～700nm)到近红外区(700～1300nm)波段,光谱反射率随病情加重呈现上升趋势,可见光520～680nm波段范围内尤为明显。当病叶严重度达到b2(25%)时,可作为病害识别的临界,对其进行早期诊断。对光谱一阶微分特征研究表明,在红边范围内(680～780nm)处理间变幅最大,分析后发现红边斜率减小,红边位置发生了"蓝移",表现出了病害特有的特征。试验证明:434～724和909～1600nm为棉花黄萎病病叶光谱敏感波段,建立的多个遥感估测模型均达到极显著水平。【结论】棉花黄萎病病叶光谱特征明显,建立的相应病害反演模型中利用一阶微分光谱723nm建立的模型精度最高,可用来定量反演棉花单叶黄萎病的发生情况。

【Objective】The aims are elucidation of spectrum characteristics of cotton leaf infected by verticillium wilt and estimation of its SL ( severity level ) and providing theoretic foundation for further monitoring cotton verticillium wilt at large scale using airborne and airspace remote sensing. 【Method】The spectrum reflectance of cotton single leaf infected by verticillium wilt was measured in cotton disease nursery and field in different growth phases, in the meanwhile, SL of single leaf infected by verticillium wilt was investigated. The method of first derivative spectrum was used to estimate accurately the disease of cotton single leaves infected by compared with the reflectance spectrum of different single leaves infected by verticillium wilt. 【Result】 The results indicated that spectrum characteristics of cotton leaf infected by verticillium wilt had better regularity with the increase of SL in different periods and varieties. Spectrum reflectance increased significantly in visible light region (400-700 nm) and near-infrared region (700-1 300 nm) with the increase of the SL, and especially significant in visible light regions (525-680 nm). When SL attained b2 (25%), cotton leaf infected by verticillium wilt could be used as a threshold and diagnose index in early period. There were evident different characteristics of first derivative spectra in these diseased leaves, it changed significantly in red edge ranges (680-780 nm) in different SL, red edge of first derivative spectrum swing decreased, and red edge position equal moved to the blue. The results indicated that 434-724 nm and 909-1 600 nm should be selected out as sensitive bands region to SLof single leaf. Some inversion models for estimating cotton leaf SL of verticillium wilt all reached the best significant level. 【Conclusion】The results suggested that different spectrum characteristics of cotton leaf infected by verticillium wilt were marked. The model in which the first derivative spectra at 723 nm could invert accurately the cotton leaf SL, and it may be used to forecast the position of cotton leaf infected by verticillium wilt quantitatively.