粒子群优化的最小二乘支持向量机在小麦白粉病监测中的应用

Application of PSO-LSSVM in Wheat Powdery Mildew Monitoring

小麦白粉病严重影响小麦的产量和品质,区域尺度上准确、及时地监测小麦白粉病的发生情况有利于高效地指导防治工作。利用Landsat-8遥感影像,提取出对小麦白粉病病情影响较大的长势因子和生境因子,包括归一化植被指数(NDVI)、比例植被指数(RVI)、绿度(Greenness)、湿度(Wetness)和地表温度(LST),利用最小二乘支持向量机(LSSVM)对陕西省关中平原部分地区的小麦白粉病进行监测,并用粒子群算法(PSO)优化模型参数,将监测结果分别与传统支持向量机(SVM)和最小二乘支持向量机(LSSVM)的监测结果进行对比分析。结果表明:经过粒子群优化的最小二乘支持向量机模型(PSO-LSSVM)的总体监测精度达到92.8%,优于SVM的71.4%和LSSVM的85.7%,取得了较好的监测效果。

The yield and quality of wheat are seriously affected by wheat powdery mildew.Accurately and timely monitoring of the occurrence of wheat powdery mildew at a regional scale can effectively guide the prevention and control work.By using Landsat-8images,this paper extracts wheat growth factors and habitat factors of affecting wheat powdery mildew,including Normalized Difference Vegetation Index（NDVI）,Ratio Vegetation Index（RVI）,greenness,wetness and Land Surface Temperature（LST）.Least Squares Support Vector Machine（LSSVM）is used to monitor the wheat powdery mildew in parts of Guanzhong plain,Shanxi Province.And the parameters of LSSVM model are optimized by Particle Swarm Optimization（PSO）algorithm.The monitoring results are compared with the traditional support vector machine（SVM）algorithm and LSSVM algorithm.The results show that the overall monitoring accuracy of the least squares support vector machine model optimized by the particle swarm optimization algorithm（PSO-LSSVM）is92.8%,which is superior to those of the traditional SVM model（71.4%）and LSSVM model（85.7%）.The proposed method obtains good monitoring results.