{"title":"Detection of Insect Invasion Symptoms on Tree Leaves Using Image Processing","authors":"Muhammad Badrisya Nordin, S. B. Hisham","doi":"10.1109/ICOCO56118.2022.10031739","DOIUrl":null,"url":null,"abstract":"This project aims to help farmers in Lumut, Perak to combat thrips invasion on mango trees. It would help reduce loss of fruit-producing branches, manual inspections, and the need to cover large acres of land manually. Data was collected by using a Canon DSLR camera at lm distance in natural lighting and uncontrolled background. Images of healthy and diseased new leaves are pre-processed to remove noise. Masking and thresholding using a range of intensity values are used to remove the background. After that, the images were clustered using Fuzzy C-Means clustering. It was found that this method was more suitable than K-Means clustering as it uses a soft clustering approach. The images obtained were then classified using Support Vector Machine (SVM). An average classification accuracy of 9S.52% was achieved.","PeriodicalId":319652,"journal":{"name":"2022 IEEE International Conference on Computing (ICOCO)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE International Conference on Computing (ICOCO)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICOCO56118.2022.10031739","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
This project aims to help farmers in Lumut, Perak to combat thrips invasion on mango trees. It would help reduce loss of fruit-producing branches, manual inspections, and the need to cover large acres of land manually. Data was collected by using a Canon DSLR camera at lm distance in natural lighting and uncontrolled background. Images of healthy and diseased new leaves are pre-processed to remove noise. Masking and thresholding using a range of intensity values are used to remove the background. After that, the images were clustered using Fuzzy C-Means clustering. It was found that this method was more suitable than K-Means clustering as it uses a soft clustering approach. The images obtained were then classified using Support Vector Machine (SVM). An average classification accuracy of 9S.52% was achieved.