园林植物常见病虫害识别  

The Identification for Common Plant Diseases and Pests from Garden Plants

作　　者：杨庆贺 丛晓燕[2] 秦丽红 郭秀梅[2] YANG Qing-he;CONG Xiao-yan;QIN Li-hong;GUO Xiu-mei(Logistics Management Office,Shandong Agricultural University,Tai'an 2710108,China;College of Information Science and Engineering/Shandong Agricultural University,Tai'an 271018,China;Tai'an Planning research Center,Tai'an 271000,China)

机构地区：[1]山东农业大学后勤管理处,山东泰安271010 [2]山东农业大学信息科学与工程学院,山东泰安271018 [3]泰安市规划编制研究中心,山东泰安271000

出　　处：《山东农业大学学报（自然科学版）》2022年第2期265-270,共6页Journal of Shandong Agricultural University：Natural Science Edition

摘　　要：绿化植物养护是校园生态文明建设的重要组成部分,在美化校园环境,愉悦师生身心健康方面起着非常重要的作用。但是随着全球气候变暖环境恶化,导致虫害和疾病经常爆发。病虫害对校园绿化防护具有极大的危害,直接影响校园绿化环境。本文针对校园绿化中常见的病虫害,提出了基于深度学习的病虫害识别方法。该方法首先对图像进行灰度归一化处理,然后提取灰度图像的PCANet特征,最后利用支持向量机进行识别。该方法在扩充的数据集上进行了验证,识别效果理想。Campus landscaping plants are important parts of campus ecological civilization construction.It plays a very important role in greening the campus environment and pleasing the physical and mental health of teachers and students.But as the global environment deteriorates,there are frequent outbreaks of plant pests and diseases.Plant diseases and pests can bring great harm to campus greening protection and directly affect the campus greening environment.In this paper,we proposed a method for plant disease and pests identification based on deep learning.Firstly,transform the color image into gray image and normalized the gray image.Secondly,extract PCANet feature of the gray image,and then use the the support vector machine to identify.The method is tested on the extended database,and the recognition is effetely.

关 键 词：园林植物 病虫害识别 深度学习 

分 类 号：S43[农业科学—农业昆虫与害虫防治]