项目名称: 基于近地面高光谱影像特征与物联网架构的松材线虫病早期智能监测研究
项目编号: No.31470579
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 农业科学
项目作者: 潘洁
作者单位: 南京林业大学
项目金额: 83万元
中文摘要: 高光谱遥感以其强大的光谱敏感性为森林病虫害早期监测预警提供了可能。本研究拟通过定点定株松材线虫接种、松枯梢菌接种及干旱胁迫实验进行近地面连续高光谱影像测量,提取不同感病时期影像特征,解决不同松树种不同胁迫类型能否根据高光谱影像特征实现早期判别以及松材线虫病感病程度的定量化预测问题;在此基础上提取松材线虫病早期诊断特征影像波段制作成具备成像功能的便携传感器,并配置GPS定位与通迅功能,使巡林员通过简单的拍照操作,相应的携带定位信息的专用诊断波段影像即时通过无线传输网络传至控制中心,通过智能监测系统软件平台进行模型分析与诊断,确认林株是否感染松材线虫病及感病程度等信息,并将此信息即时传回给巡林员,其实时采取相应的管理调控措施。如此构建起基于物联网架构的便携式松材线虫病早期智能监测系统,将实现符合实际应用需求的实时监测功能,有效地提高森林松材线虫病的防治效率,为进行科学决策和快速反应提供了可能。
中文关键词: 松材线虫病;早期;智能监测;高光谱影像;物联网
英文摘要: Hyperspectral remote sensing with its powerful spectral sensitivity provides possibility for the early monitoring on forest diseases and pests. This research proposes to conduct continuous measurements of ground hyperspectral images by the fixed location and fixed plants experiments: the pine wood nematode inoculation, pine shoot blight bacteria inoculation and drought stress experiments, to obtain hyperspectral imaging features in different disease or stress periods and to resolve the problems such as: if the hyperspectral imaging features can realize the early distinguish among different pine species and different stress types or not? if the pine wilt disease degree can be quantified by hyperspectral imageing features or not? Based on the solution of these problems, the specific hyperspectral image bands or bands combination for early monitoring on the pine wilt disease will be used to produce the portable sensor with imaging, GPS location and communication function. This portable sensor can be used by the forest rangers to simply take some photos on the pine plants. These special photos will be immediately transmitted to the control center by wireless transmission network and then be analyzed by models in the software platform of the intelligent monitoring system to confirm the pine plants infected by pine wilt disease or not and to quantify the infection degree if its infection are confirmed. The confirm information will turn back to the portable sensor to instruct the forest rangers managing these pine plants in time. So, a portable intelligent system for early pine wilt disease monitoring based on the internet of things framework will be established. Successful completion of this proposed research will provide a method to meet the practical application demand of real-time monitoring function, effectively improve the prevention efficiency of forest pine wilt disease and provide the possibility of the scientific decision-making and rapid response to the pine wilt disease.
英文关键词: Pine wilt disease;early;intelligent monitoring;hyperspectral images;internet of things