View Item 
      •   UMY Repository
      • 01. BOOKS
      • Books
      • View Item
      •   UMY Repository
      • 01. BOOKS
      • Books
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Machine Learnings of Dental Caries Images based on Hu Moment Invariants Features

      Thumbnail
      View/Open
      Artikel (466.5Kb)
      Hasil turnitin (1.229Mb)
      Date
      2021
      Author
      Jusman, Yessi
      Anam, Muhammad Khoirul
      Puspita, Sartika
      Saleh, Edwyn
      Metadata
      Show full item record
      Abstract
      Dental caries generally occurs due to consuming foods containing carbohydrates, such as sucrose, and rarely brushing teeth, thus causing gradual damage to the layers and structures of the teeth. This study aims to build a dental caries level classification system using image processing and machine learning methods. The first step was to analyze and discover the extraction results from Hu’s moment invariants. After successfully extracting the features, the classification was carried out using a Support Vector Machine (SVM) and K- Nearest Neighbors (KNN). This study employed radiographic images of four dental caries classes consisting of Class 1, 2, 3, and 4. A total of 198 images of dental caries were used as training data and 66 images as test data. The classification obtained accuracy value of the SVM and KNN. The highest accuracy was discovered in the Fine Gaussian model of the SVM classification method with 77.6%, while the lowest accuracy was depicted in the Cubic model with 57.4%. Meanwhile, the highest accuracy by using KNN is 100% of accuracy using Fine and Weighted KNN models.
      URI
      http://repository.umy.ac.id/handle/123456789/36583
      Collections
      • Books

      DSpace software copyright © 2002-2015  DuraSpace
      Contact Us | Send Feedback
      Theme by 
      @mire NV
       

       

      Browse

      All of UMY RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      DSpace software copyright © 2002-2015  DuraSpace
      Contact Us | Send Feedback
      Theme by 
      @mire NV