NPS logo Naval Postgraduate School
Dudley Knox Library
        View Item 
        •   Calhoun Home
        • Theses and Dissertations
        • 1. Thesis and Dissertation Collection, all items
        • View Item
        •   Calhoun Home
        • Theses and Dissertations
        • 1. Thesis and Dissertation Collection, all items
        • View Item
        • How to search in Calhoun
        • My Accounts
        • Ask a Librarian
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of CalhounCollectionsThis Collection

        My Account

        LoginRegister

        Statistics

        Most Popular ItemsStatistics by CountryMost Popular Authors

        Map classification in image data

        Thumbnail
        View/Open
        Icon15Sep_Fiebiger_Frank.pdf (1.502Mb)
        Download Record
        Download to EndNote/RefMan (RIS)
        Download to BibTex
        Author
        Fiebiger, Frank
        Date
        2015-09
        Advisor
        Kölsch, Mathias N.
        Second Reader
        Buttrey, Samuel E.
        Metadata
        Show full item record
        Abstract
        The digital era has led to an unprecedented increase in the amount of information available, of which an essential part is represented by visual data. The data forensics community asks for machine solutions to face the proliferation of image data. This thesis addresses the specific problem of distinguishing two-dimensional map images from other image content by examining two computational meth-ods: Convolutional Neural Networks (CNNs) and Bag of Words (BOW). No information about current automated solutions for the mentioned task is available. The CNN used in this research consists of 60 million parameters and 650,000 neurons in eight weighted layers, is pre-trained on 1,000 classes, and provides an immense learning capacity. The BOW method uses a visual vocabulary, con-structed by clustering higher-level image information, to classify unknown images by comparing their contained visual words with a content-specific vocabulary of a classifier. Both methods are evaluated in terms of recall and precision, or percentage of correctly and incorrectly classified images. The data collection consists of 1,200 map images called positives, subdivided into four sub-classes, and an additional 1,200 images without map content, called negatives. Results with a recall up to 99.17%and corresponding precision up to 97.01% support the idea of implementing CNN and BOW as the backbone of a computer-based classification application.
        Description
        Approved for public release; distribution is unlimited
        URI
        http://hdl.handle.net/10945/47255
        Collections
        • 1. Thesis and Dissertation Collection, all items

        Related items

        Showing items related by title, author, creator and subject.

        • Thumbnail

          Variable-retardance, Fourier-transform imaging spectropolarimeters for visible spectrum remote sensing 

          Tyo, J. Scott; Turner, Theodore S. (2001-03);
          An imaging, variable-retardance, Fourier-transform spectropolarimeter is presented that is capable of creating spectropolarimetric images of scenes with independent characterization of spatial, spectral, and polarimetric ...
        • Thumbnail

          Reconstruction of computer simulated, atmospheric turbulence-degraded, astronomical objects by application of the Knox-Thompson and triple-correlation phase recovery techniques 

          Lackemacher, James M. (Monterey, California: Naval Postgraduate School, 1990-12);
          Atmospheric turbulence severely degrades images of astronomical objects. Providing images that accurately reflect the true nature of these objects is essential to their understanding. Several object recovery techniques ...
        • Thumbnail

          Precise and Efficient Retrieval of Captioned Images: The MARIE Project 

          Rowe, Neil C. (Monterey, California. Naval Postgraduate School, 1999);
          The MARIE project has explored knowledge-based information retrieval of captioned images of the kind found in picture libraries and on the Internet. It exploits the idea that images are easier to understand with context, ...
        Feedback

        411 Dyer Rd. Bldg. 339
        Monterey, CA 93943

         

        circdesk@nps.edu
        (831) 656-2947
        DSN 756-2947

        Start Your Research

        • Research Guides
        • How to Cite
        • Search Basics
        • Ask a Librarian
        • Library Liaisons
        • Graduate Writing Center
        • Thesis Processing Office
        • Statistics, Maps & More
        • Copyright at NPS

        Find & Download

        • Databases List
        • Articles, Books & More
        • NPS Theses
        • NPS Faculty Publications: Calhoun
        • Journal Titles
        • Course Reserves

        Use the Library

        • My Accounts
        • Request Article or Book
        • Borrow, Renew, Return
        • Remote Access
        • Workshops & Tours
        • For Faculty & Researchers
        • For International Students
        • For Alumni
        • Print, Copy, Scan, Fax
        • Rooms & Study Spaces
        • Floor Map
        • Computers & Software
        • Adapters, Lockers & More

        Collections

        • NPS Archive: Calhoun
        • Restricted Resources
        • Special Collections & Archives
        • Federal Depository
        • Homeland Security Digital Library

        About

        • Hours
        • Library Staff
        • About Us
        • Visit Us

        NPS-Licensed Resources - Terms & Conditions

        Copyright Notice

         
         

        Facebook logo Federal Depository Library Program Emblem NPS Video Portal

        NPS Home Privacy Policy Copyright Accessibility Contact Webmaster