• DocumentCode
    2252258
  • Title

    Accurate detection of demosaicing regularity from output images

  • Author

    Cao, Hong ; Kot, Alex C.

  • Author_Institution
    Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore, Singapore
  • fYear
    2009
  • fDate
    24-27 May 2009
  • Firstpage
    497
  • Lastpage
    500
  • Abstract
    Demosaicing regularity is an important processing regularity associated with the internal camera processing and its detection from output photos is useful for non-intrusive forensic engineering. In this paper, we propose a reverse grouping technique to improve the detection accuracy of our earlier proposed detection model based on second-order image derivatives. Comparison results based on syntactic images shows that the proposed technique significantly reduces the reprediction errors for some commonly used demosaicing algorithms. When applied to a real application, i.e. camera model identification, our demosaicing features in conjunction with probabilistic support vector machine classifier achieve excellent classification performance.
  • Keywords
    image classification; image reconstruction; image segmentation; probability; security of data; support vector machines; camera model identification; demosaicing regularity detection; image reconstruction; internal camera processing; nonintrusive forensic engineering; output image; probabilistic support vector machine classifier; reverse grouping technique; second-order image derivative; syntactic image; Digital cameras; Digital images; Filtering; Forensics; Image coding; Optical distortion; Optical filters; Sensor phenomena and characterization; Transform coding; Watermarking;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems, 2009. ISCAS 2009. IEEE International Symposium on
  • Conference_Location
    Taipei
  • Print_ISBN
    978-1-4244-3827-3
  • Electronic_ISBN
    978-1-4244-3828-0
  • Type

    conf

  • DOI
    10.1109/ISCAS.2009.5117794
  • Filename
    5117794