Source Camera Identification for Heavily JPEG Compressed Low Resolution Still Images*

Authors

  • Erwin J. Alles M.Sc.,

    1. Department of Digital Technology and Biometry, Netherlands Forensic Institute, The Hague, Holland.
    Search for more papers by this author
    • Present Address: Faculty of Applied Sciences, Department Imaging Science and Technology, Laboratory of Acoustical Imaging and Sound Control, Delft University of Technology, Delft, Holland.

  • Zeno J. M. H. Geradts Ph.D.,

    1. Department of Digital Technology and Biometry, Netherlands Forensic Institute, The Hague, Holland.
    Search for more papers by this author
  • Cor J. Veenman Ph.D.

    1. Department of Digital Technology and Biometry, Netherlands Forensic Institute, The Hague, Holland.
    2. Intelligent Systems Lab, University of Amsterdam, Amsterdam, Holland.
    Search for more papers by this author

  • *

    A portion of this work was presented at the International Conference on Computational Science and its Applications (ICCSA) in Perugia, Italy from June 30 to July 3, 2008.

Additional information and reprint requests:
Cor J. Veenman, Ph.D.
Department of Digital Technology and Biometry
Netherlands Forensic Institute The Hague, Holland
Email: c.j.veenman@uva.nl

Abstract

Abstract:  In this research, we examined whether fixed pattern noise or more specifically Photo Response Non-Uniformity (PRNU) can be used to identify the source camera of heavily JPEG compressed digital photographs of resolution 640 × 480 pixels. We extracted PRNU patterns from both reference and questioned images using a two-dimensional Gaussian filter and compared these patterns by calculating the correlation coefficient between them. Both the closed and open-set problems were addressed, leading the problems in the closed set to high accuracies for 83% for single images and 100% for around 20 simultaneously identified questioned images. The correct source camera was chosen from a set of 38 cameras of four different types. For the open-set problem, decision levels were obtained for several numbers of simultaneously identified questioned images. The corresponding false rejection rates were unsatisfactory for single images but improved for simultaneous identification of multiple images.

Ancillary