基于高光谱遥感的小麦全蚀病监测机理研究

项目名称： 基于高光谱遥感的小麦全蚀病监测机理研究

项目编号： No.31301604

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 农业科学

项目作者： 乔红波

作者单位： 河南农业大学

项目金额： 23万元

中文摘要： 小麦全蚀病近年来在各麦区普遍严重发生，对其发生发展监测是治理的关键。本研究利用高光谱地物波谱仪和成像遥感仪，植物病理学与植物生理生化相结合的方法，研究不同抗病小麦品种在不同为害等级水平下，植株形态和生理生化参数与反射光谱信息的时空变异规律，提取小麦全蚀病高光谱遥感监测的敏感波段。利用基于支持向量机的计算机机器学习技术获取全蚀病为害光谱信息，构建基于特征光谱参数和感病等级相关的生理参数间的监测模型，提炼全蚀病遥感识别与诊断的适宜指标，并进行验证。研究结果将为小麦全蚀病高光谱监测提供理论基础和技术途径，同时对于创新作物病虫害监测方法，克服传统手段局限性，保障粮食安全生产具有重要意义。

中文关键词： 小麦全蚀病；高光谱遥感；光谱特征；支持向量机；

英文摘要： Wheat take-all is one of the devastating diseases threatening wheat production . Monitoring the disease occurrence and development is the key for disease forecast and management. The goal of this study is to develop a hyperspectral imaging approach for detecting wheat take-all disease. Hyperspectral Handheld spectrometer and imaging spectroscopy will be utilized to obtain reflectance spectral information of different wheat varieties with various disease incidences. Identification of the take-all disease will be determined by the means of plant pathology and biochemistry. The sensitive reflectance spectra of take-all infected wheat will be identified. Additionally, support vector machine-based computer machine learning techniques will be used to collect wheat take-all spectral information. A method model will be built and verified to detect take-all disease based on the relationship between disease incidence and hyperspectral data. The result of this study will not only provide theoretical and technical support for wheat no-destructive detection and safety production, but also shed light on the development of novel strategy to detect and control crop pest and disease, which has great significance to the food safety.

英文关键词： wheat take-all；hyperspectral remote sensing；spectrum；support vector machine；