机器视觉和AdaBoost的柑桔溃疡病自动检测.PDFVIP

25 7 2008 7 : 1006- 9348( 2008) 07- 0237- 03 AdaBoost 1 1 2 1 朱庆生 , 张敏 , 杨方云 , 柳锋 ( 1. , 400030; 2. , 400700) : , , Gabor, ; AdaBoost; AdaBoost 95% , 99% , , : ; ; ; : S126 : A C itrus Canker Automatic Detection Based on Com puter Vision and AdaBoostA lgorithm 1 1 2 1 HU Q ing- sheng , HANGM in , YANG Fang- yun , LIU Feng ( 1. Com puter College of Chongq ing Un iversity, Chongqing 400030, China; 2. Citrus Institute of Ch ina Agricu lture Academy, Chongq ing 400700, Ch ina) ABSTRACT: Computer vision technology is introduced into fast and accurate autom atic detection of citrus canker. Three key issues are discussed in th is paper, w hich are feature construction, feature selection and system design. To construct features, Gabor transform ation and edge detection algorithm s are used and a feature set of color, texture and shape is obtained. Then AdaBoost algorithm is applied to select the most efficient features from the feature set. Ada- Boost algorithm is also used for train ing classifier. To fast detect citrus d isease area on mi age, a movingw indow tech- nology is used. The expermi ent resu lts show that the detection accurate rate can reach over 95% , andwhen the train- ing round reachs a certain amount the detection rate can reach 99% , wh ich is about equal to the detection accurate rate by experts eyeballing, and this approach is valuable for plant production. KEYWORDS: Featu re construction; Feature selection; Classifier; Detection rate ,