基于近红外光谱番茄黄化曲叶病抗病性识别探究.docVIP

基于近红外光谱番茄黄化曲叶病抗病性识别探究 　　摘要：使用传统的聚合酶链反应技术检测番茄（Lycopersicon esculentum Mill.）植株是否有黄化曲叶病的抗病基因，进而确定植株是否具有抗病性；采集鉴定后的植株叶片的近红外漫反射光谱，采用多种方法对原始光谱进行预处理，并将不同预处理后的数据以及原始光谱作为输入，利用支持向量机建立抗病性的识别模型。采用标准正态变量变换和去趋势算法预处理后所建立的模型对预测集的识别准确率可以达到96.153 8%。表明通过近红外光谱技术可以识别番茄植株对黄化曲叶病是否具有抗病性 关键词：番茄（Lycopersicon esculentum Mill.）黄化曲叶病；抗病性；近红外光谱；光谱预处理；支持向量机 中图分类号：S123；O657.33 文献标识码：A 文章编号：0439-8114（2017）05-0953-04 DOI：10.14088/ki.issn0439-8114.2017.05.042 Study of Identifying Tomato Yellow Leaf Curl Disease Resistance Based on Near-infrared Spectroscopy Analysis Technique YANG Zeng-chong，LIU Gui-li，LI Xiang （School of Instrument Science and Opto Electronics Engineering，Beijing Information Science Technology University，Beijing 100192，China） Abstract： Gene of tomato（Lycopersicon esculentum Mill.） yellow leaf curl disease resistance has been detected by using polymerase chain reaction. This method is widely used to determine the disease resistance of tomato plants. The near-infrared diffuse reflectance spectroscopy of the identified samples was collected. Then the original data and the data preprocessed by different methods are used as the input of Support Vector Machine（SVM） to build the model to judge whether a tomato plant is resistant to the disease. Experimental results show that the model which used the correction method of Standard Normal Variate and Detrending had the best performance and the recognition accuracy of the test set can reach 96.153 8%. The result proves that it is feasible to identify the resistance of Tomato Yellow Leaf Curl Disease by Near Infrared Spectroscopy. Key words： tomato （Lycopersicon esculentum Mill.） yellow leaf curl disease；resistance；NIR；spectral pretreatment；SVM 番茄（Lycopersicon esculentum Mill.）因其?I养丰富、高产等特点已经成为中国乃至世界上的重要农作物之一。番茄黄化曲叶病（TYLCD）会使番茄幼苗生长缓慢甚至停滞，后期感染这种病也会影响果实的产量和品质，甚至造成绝收，已成为限制番茄生产的重要病害之一[1]。黄化曲叶病病毒（TYCLV）自然条件下通过烟粉虱传播和扩散，烟粉虱繁殖能力强，寄主广泛，易产生抗药性，难以从寄主方面阻止病毒的传播和扩散。再加上TYCLD易爆发、扩散迅速[2]，选取具有抗病性的番茄植株进行精细化的栽培能够保障植株后期的发育成长，保证番茄的产量和经济收益