{"title":"Color and Texture Feature based for Maize Leaf Disease Identification","authors":"I. Siradjuddin, Arif Riandika, W. Agustiono","doi":"10.1109/CENIM56801.2022.10037498","DOIUrl":null,"url":null,"abstract":"This paper presents a combination of color and texture features to identify the maize plant disease based on maize leaf image. Global Color Histogram and Color Coherence Vector extract color features, and Local Binary pattern is used to extract the texture features. There are four classes in the proposed identification model: Cercospora Leaf Spot, Common Rust, healthy, and Northern Leaf Blight disease. For the identification process, we trained the Voting classifier with five CARTs and a plant-village dataset. The trained identification model achieved an average accuracy is 75,935%. In addition, we added a segmentation image for the preprocessing stage to improve the accuracy. As a result, this preprocessing stage increased the accuracy to 82.645%.","PeriodicalId":118934,"journal":{"name":"2022 International Conference on Computer Engineering, Network, and Intelligent Multimedia (CENIM)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Computer Engineering, Network, and Intelligent Multimedia (CENIM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CENIM56801.2022.10037498","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
This paper presents a combination of color and texture features to identify the maize plant disease based on maize leaf image. Global Color Histogram and Color Coherence Vector extract color features, and Local Binary pattern is used to extract the texture features. There are four classes in the proposed identification model: Cercospora Leaf Spot, Common Rust, healthy, and Northern Leaf Blight disease. For the identification process, we trained the Voting classifier with five CARTs and a plant-village dataset. The trained identification model achieved an average accuracy is 75,935%. In addition, we added a segmentation image for the preprocessing stage to improve the accuracy. As a result, this preprocessing stage increased the accuracy to 82.645%.