EEPIS Repository

PENGELOMPOKAN GAMBAR BERDASARKAN WARNA DAN BENTUK MENGGUNAKAN FGKA (FAST GENETIC KMEANS ALGORITHM) UNTUK PENCOCOKAN GAMBAR

RAHMANTI , FARAH ZAKIYAH and Martiana, Entin and Ramadijanti , Nana PENGELOMPOKAN GAMBAR BERDASARKAN WARNA DAN BENTUK MENGGUNAKAN FGKA (FAST GENETIC KMEANS ALGORITHM) UNTUK PENCOCOKAN GAMBAR. EEPIS Final Project.

[img]
Preview
PDF - Published Version
Download (190Kb) | Preview

    Abstract

    A large Collection of digital images in many areas of aspect are being created. The collection images are digitizing result of analogue photographs, diagrams, paintings, drawings, prints. Usually,the way of searching these collections was by indexing and image information based on text (like caption or keywords). This way is not effective and efficient because two reasons, are big size of database and subjective in picture meaning. Recently, it has been developed many ways in image retrieval that use image content (color, shape, and texture). The use of centroid produced from clustered RGB Histogram and Edge Detection matrix using FGKA, can be used for searching parameter. FGKA is merger of Genetic Algorithm and Kmeans Clustering Algorithm. FGKA is also developed from Genetic Kmeans Algorithm (GKA) which is always converge to global optimum. Image Clustering and Matching based on color-shape feature are better than based on color feature only if using some data wich are greatly into shape feature.

    Item Type: Article
    Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
    Q Science > QA Mathematics > QA76 Computer software
    Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
    Depositing User: Mrs Ashri Esy
    Date Deposited: 29 Jul 2011 15:05
    Last Modified: 29 Jul 2011 15:05
    URI: http://repo.eepis-its.edu/id/eprint/720

    Actions (login required)

    View Item