| 摘要: |
| 【目的】 分析红富士苹果树分别在各级红蜘蛛虫害、黄叶病害胁迫下的反射光谱特征，并利用光谱数据对果树受害程度进行定量化测评。【方法】 于2007年6~7月，在陕西省礼泉县测定了患有各级红蜘蛛虫害、黄叶病害果树的光谱反射率，将正常、轻度、中度、重度受害果树的光谱反射率分别记为R正常、R轻度、R中度、R重度。首先，在对光谱数据进行预处理和一阶微分变换的基础上，对各级红蜘蛛虫害、黄叶病害胁迫下果树的光谱特征分别进行分析；随后，基于各级红蜘蛛虫害、黄叶病害对应的光谱特征，分别构造了6种光谱指数，并从中择优建立对红蜘蛛虫害、黄叶病害级别进行测评的数学模型；最后，利用检验样本果树的反射光谱分别对所建立的红蜘蛛虫害、黄叶病害级别测评模型的精度进行检验。【结果】 各级红蜘蛛虫害对应的反射光谱在630～695 nm，R重度＞R中度＞R轻度＞R正常，且在684 nm处变异系数最大；在730～950 nm，R重度＜R中度＜R轻度＜R正常，且在762 nm处变异系数最大。各级黄叶病害对应的反射光谱在515～716 nm，R重度＞R中度＞R轻度＞R正常，且在603 nm处变异系数最大；在740～950 nm，R重度＜R中度＜R轻度＜R正常，且在764 nm处变异系数最大。果树反射光谱的“红边”位置随虫害、病害程度的加重而依次发生“蓝移”。建立的红蜘蛛虫害、黄叶病害级别测评模型对相应虫害、病害级别的测评准确率分别为96%，98%。【结论】 各级红蜘蛛虫害、黄叶病害胁迫下的红富士苹果树的反射光谱差异较大，利用光谱数据可精确地对各级红蜘蛛虫害、黄叶病害级别进行定量化测评。 |
| 关键词: 苹果树 反射光谱 黄叶病害 红蜘蛛虫害 定量化测评 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目（30571527） |
|
| Quantiative evaluation of the degrees of disease or insect pest intimidation based on spectral reflectance of canopies of fruit trees |
|
LI Dan dan SI Long ting LUO Xiao mei LI Tao
|
| Abstract: |
| 【Objective】 The research attempts to explore the spectral characteristics of FuJi apple trees which are intimidated by disease or pests,and try to evaluate quantitatively the degrees of disease or pest stress by spectra data.【Method】 During June and July 2007,the reflectance spectroscopies of the FuJi apple trees intimidated by each level of disease or pests were measured in liquan county,and the spectral reflectances were marked respectively:Rsevere,Rmoderate,Rlightly,Rnormal.Firstly,The spectra reflectance (Rλ) data were pretreated and transformed with the first derivative,and on the basis of it,the spectral characteristics of fruit trees intimidated by all levels of disease or insect pest were analyzed separately.Subsequently,on the basis of the spectral characteristics of fruit trees intimidated by each level of disease or pests,six kinds of spectra index were conformed separately,and the optimal spectra index was selected and used to model each level of disease or pests.Finally,the models that were used to evaluate quantitatively the degrees of disease or pest stress were checked up separately.【Result】 The main conclusions are: (1) The Spectral reflectance of each level of red mite pests within the wavelen gth ranges from 630 to 695 nm,Rsevere>Rmoderate>Rlightly>Rnormal，the largest coefficient of variation of the four reflectances is at 684 nm,within the wavelength range from 730 to 950 nm,Rseveremoderatelightlynormal，the largest coefficient of variation of the four reflectances is at 762 nm.The spectral reflectance of each level of yellow leaves disease within the wavelength ranges from 515 to 716 nm,Rsevere>Rmoderate>Rlightly>Rnormal，the largest coefficients of variation of the four reflectances is at 603 nm,within the wavelength range from 740 to 950 nm,Rsevere<Rmoderatelightlynormal，the largest coefficient of variation of the four reflectances is at 764 nm.(2) The red border position of the spectral reflectance of fruit trees moves to shortwave with the level of disease or pest increasing.(3)The models that are used to evaluate quantitatively the degrees of disease or pest stress have higher precision.【Conclusion】 The discrepancy among the reflectance spectroscopy of each level of disease or pest intimidating the FuJi apple trees is larger,the precision using the spectra data to evaluate quantitatively the degree of disease or insect pest intimidation is high. |
| Key words: apple tree spectral reflectance yellow-leaf disease european red mite pest quantitative evaluation |
