摘要
马尾松毛虫是马尾松林的主要害虫,危害很大,实施有效监测是对其进行防治的关键。以福建沙县为研究区,以SPOT-5遥感影像为数据源,基于短波红外波段、红光波段和近红外波段等构造归一化植被指数等光谱特性指数,基于灰度共生矩阵法提取均值等影像纹理特征,引入坡度、坡向等地形及林木因子构建马尾松毛虫虫害遥感监测指标,通过岭回归分析建立虫情级数估测模型进而反演,进行虫害监测信息提取,并与单纯基于光谱特性指数指标的虫害信息提取方法进行比较。结果表明:1)综合考虑影像的光谱特性、纹理特征与地形及林木因子影响进行虫害监测,显著增强了虫害的光谱响应能力,相比于单纯基于光谱特性的虫害监测信息提取,其总精度提高了14.28%;2)这些为林业生产实践提供技术支撑与理论借鉴,同时也在一定程度上推动遥感监测森林病虫害的理论认识与信息提取方法。
Dendrolimus punctatus is a major pest of Pinus massoniana forests and is very harmful, effective monitoring is the key step for controlling Dendrolimus punctatus. Shaxian county of Fujian Province was chosen as the study area and SPOT-5 Remote Sensing Image as the data sources. Normalized vegetation indexes (other spectral characteristics indexes) were generated based on shortwave infrared band, red band and near infrared band. The image texture features such as mean values were extracted based on gray level cooccurrence matrix method. The remote sensing monitoring index system of D.punctatus insect pests was constructed by introducing topographic factors such as slope, slope direction and forest tree factors. And the insect pest level estimation model was established by ridge regression analysis, and then the inversion was carried out. With this model, the insect pest monitoring information was extracted and compared with that of the pest information extraction method based solely on spectral characteristic index. The results show that 1) the method of comprehensively considering spectrum characteristic, texture features, terrain and forest factors had better effects than the method only based on spectrum characteristic, which highly enhanced the D. punctatus damage spectral response and the total accuracy was Increased by 14.28%;2) These methods provide a technical support and theoretical reference for forestry production practices, and at the same time promote the theoretical understanding and information extraction methods of remote sensing monitoring of forest pests and diseases.
作者
亓兴兰
肖丰庆
刘健
张李平
QI Xinglan;XIAO Fengqing;LIU Jian;ZHANG Liping(Fujian Forestry Vocational Technical College,Nanping 353000,Fujian,China;Nanping Agriculture Bureau,Nanping 353000,Fujian China;College of Forestry,Fujian Agriculture and Forestry University,Fuzhou 350002,Fujian,China;Key Laboratory of 3S Technology and Resources Optimization of Fujian Provincial University,Fuzhou 350002,Fujian,China)
出处
《中南林业科技大学学报》
CAS
CSCD
北大核心
2019年第4期59-65,共7页
Journal of Central South University of Forestry & Technology
基金
国家自然科学基金项目(30871965)
福建省自然科学基金项目(2016J05072)
福建省中青年教师教育科研项目(JAT160744
JA14390)
关键词
马尾松毛虫
虫害遥感监测
SPOT-5遥感影像
福建沙县
Dendrolimus punctatus
Monitor
monitoring insect pest by remote sensing
SPOT-5 remote sensing image
Shaxian County, Fujian province
