基于高光谱成像的寒地水稻叶瘟病与缺氮识别

发布时间：2018-06-18 17:15

本文选题：光谱分析 + 水稻叶瘟病　；参考：《农业工程学报》2016年13期

【摘要】：为进行水稻叶瘟病与养分缺失的区分、实现叶瘟病及时、准确的诊断,以大田试验为基础,利用高光谱成像仪获取2个品种的健康、缺氮、轻度感病和重度感病共4类水稻叶片的反射率光谱,对其光谱特性进行分析,并采用多种预处理方法、分别结合偏最小二乘判别分析(partial least squares-discriminate analysis,PLS-DA)和主成分加支持向量机(principle component analysis-support vector machine,PCA-SVM)方法构建水稻叶瘟病识别模型。试验结果显示6个判别模型都获得了较高的识别准确率,经标准正态变量(standard normal variate,SNV)变换预处理的PLS-DA模型获得了最佳的识别结果,预测准确率达100%,经多元散射校正(multiplicative scatter correction,MSC)预处理的PCA-SVM模型的预测准确率也达到97.5%。本研究为水稻叶瘟病的判别和分级提供了新方法,也为稻瘟病大范围遥感监测提供了基础。
[Abstract]:In order to distinguish rice leaf blast from nutrient deficiency and realize timely and accurate diagnosis of rice leaf blast, based on field experiments, two varieties were obtained by using hyperspectral imager. The spectral characteristics of four kinds of rice leaf reflectance spectrum were analyzed, and various pretreatment methods were used. Combined with partial least squares-discriminate analysis (PLS-DA) and principal component plus support vector machine (principle component analysis-support vector machine PCA-SVM), a rice leaf blast identification model was constructed. The experimental results show that all the six discriminant models have high recognition accuracy, and the PLS-DA model preprocessed by standard normal variable SNV transformation has the best recognition result. The prediction accuracy of PCA-SVM model pretreated by multiple scattering correction scatter correction MSC( MSC-PCA-SVM) is up to 97.5%. This study provides a new method for the identification and classification of rice leaf blast and provides a basis for large-scale remote sensing monitoring of rice blast.
【作者单位】：东北农业大学电气与信息学院;黑龙江东方学院计算机科学与电气工程学部;
【基金】：国家“863”计划项目(2013AA102303) 黑龙江省重大科技研发项目(GY2014ZB0011) 哈尔滨市科技攻关项目(2013AA6BN010)
【分类号】：S435.111.41;TP751

【相似文献】