• 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.

    Pengenalan Gerak Isyarat Menggunakan Layar Sentuh Virtual dan Neural Network Backpropagation

    View/Open
    Fuultext (4.400Mb)
    Date
    2014
    Author
    Priyowidodo, Stephanus
    Advisor(s)
    Zarlis, Muhammad
    Sitompul, Opim Salim
    Metadata
    Show full item record
    Abstract
    Gesture recognition is very important to use several types of applications such as human machine interface, control robot interaction, robot for personal use, special monitoring and robotic systems as a special tool. Constraints in gesture recognition techniques including retrieval object coordinates, object data processing, the selection of the neural network parameters and the level of recognition accuracy. This study uses a virtual touch screen technique to obtain the coordinates of the data object, the data is further processed by the Bresenham algorithm and used as inputs to the neural network backpropagation. The results of this study indicate that the virtual touch screen can be used as an alternative to the object coordinate data collection, Bresenham algorithm can help maximize the depiction of data and back propagation neural network can recognize all gestures or one hundred percent of the data well after the error is smaller 0.8.
     
    Pengenalan gerak isyarat sangat penting untuk penggunaan beberapa jenis aplikasi seperti antar-muka manusia mesin, pengendali interaksi robot, robot untuk keperluan pribadi, sistem pemantau khusus dan robot sebagai alat bantu khusus. Kendala dalam pengenalan gerak isyarat diantaranya teknik pengambilan koordinat objek, pemrosesan data objek, pemilihan parameter neural network dan tingkat ketepatan pengenalan. Penelitian ini menggunakan teknik layar sentuh virtual untuk memperoleh data koordinat objek, data selanjutnya diproses dengan algoritma Bresenham dan dijadikan input pada neural network backpropagation. Hasil penelitian ini menunjukkan bahwa layar sentuh virtual dapat digunakan sebagai alternatif pengambilan data koordinat objek, algoritma Bresenham dapat membantu memaksimalkan penggambaran data dan neural network backpropagation dapat mengenali seluruh gerak isyarat atau seratus persen data dengan baik setelah error lebih kecil 0.8.

    URI
    http://repositori.usu.ac.id/handle/123456789/41312
    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