• Login
    View Item 
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Temu Kembali Citra Wajah Berdasarkan Pengukuran Kemiripan Fitur dengan Menggunakan Jaringan Bayesian

    View/Open
    Fulltext (4.049Mb)
    Date
    2013
    Author
    Siagian, Hendrik
    Advisor(s)
    Sihombing, Poltak
    Zarlis, Muhammad
    Metadata
    Show full item record
    Abstract
    In this study, the characteristics of the face image is expressed through three image features extracted directly from the key facial image color features, shape features and texture features. Color feature extracted by using color histograms HSI (hue, saturation, intensity); shape features extracted by using Sobel operator and arranged in edge direction histogram; texture features extracted by using co-occurence matrix. Characteristics of the query image and the images in the database can be considered as the nodes that are interconnected and form a Bayesian network. Bayesian network is a graph illustrating the structure of relationships among chance variables in a large number of exciting opportunities and inference on the set of variables. Link between two variables or nodes will represent opportunities occurrence of the degree of similarity with the query image of each image in the database, can be measured by comparing the query image characteristics with the characteristics of the images in the database. The evaluation of the results of image retrieval precision for each recall faces show very good performance of Bayesian network.
     
    Dalam penelitian ini, karakteristik citra wajah dinyatakan melalui tiga buah fitur citra yang diekstrak secara langsung dari citra wajah kunci yaitu fitur warna, fitur bentuk dan fitur tekstur. Fitur warna diekstraksi dengan menggunakan histogram warna HSI (hue, saturation, intensity); fitur bentuk diekstraksi dengan menggunakan operator Sobel dan disusun dalam edge direction histogram; fitur tekstur diekstraksi dengan menggunakan co-occurence matrix. Karakteristik citra query dan citra-citra yang ada di dalam database dapat dianggap sebagai node-node yang saling berhubungan dan membentuk sebuah jaringan Bayesian. Jaringan Bayesian merupakan struktur grafik yang menggambarkan peluang relasi diantara variabel-variabel dalam jumlah yang besar dan dapat menarik peluang inferensi atas variabel-variable tersebut. Link antara dua variabel atau node akan merepresentasikan peluang kejadian dari derajat kemiripan citra query dengan setiap citra dalam database dapat diukur dengan cara membandingkan karakteristik citra query dengan karakteristik citra-citra dalam database. Evaluasi terhadap precision hasil temu kembali citra wajah untuk setiap recall memperlihatkan kinerja jaringan Bayesian sangat baik.

    URI
    http://repositori.usu.ac.id/handle/123456789/40539
    Collections
    • Master Theses [621]

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of USU-IRCommunities & CollectionsBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit DateThis CollectionBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit Date

    My Account

    LoginRegister

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV