Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Journal of Chinese Agricultural Mechanization》 2017-04
Add to Favorite Get Latest Update

Identification method of grape diseases based on computer technology

Liu Yuan;Feng Quan;Gansu Agricultural University;  
Diseases restrict the growth and health of grape,and affect the yield and quality of it,which make detection and diagnosis be the prerequisite of disease control.Aiming at the features of diseased grape leaves images,recognizing method for grape disease is studied in this paper with computer image processing and pattern identification,which can improve the accuracy and efficiency of recognition.Firstly,the diseased leaves images in different growth stages with natural illumination were collected,thus,28 kinds of feature vectors from diseased parts were extracted,including color and texture.Secondly,in order to choose the best classifier for recognizing 6 common grape diseases,22 classifiers(Decision Tree,Discriminant Analysis,SVM,NN,Ensamble Classifier,etc.)were trained and tested.Based on that,Bagged Trees is chosen to be the most accurate classifier,the recognition rate is 86.4%,and rate of recognizing powdery mildew,brown spot,downy mildew,gray mold,black pox and anthracnose are 93.0%,87.5%,82.7%,87.7%,85.5% and 82.1% respectively.Finally,the test results indicate that this method can effectively recognizing several diseases of grape leafs,and provide method and references for rapid and automatic recognition of grade diseases.
【Fund】: 国家自然科学基金(61461005);; 甘肃省科技重大专项计划(1502NKDF023)
【CateGory Index】: S436.631;TP391.41
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved