Cover Image
QR Code

Image re-sampling detection through a novel interpolation kernel.

Image re-sampling involved in re-size and rotation transformations is an essential element block in a typical digital image alteration. Fortunately, traces left from such processes are detectable, proving that the image has gone a re-sampling transformation. Within this context, we present in this p...

Full description

Published in:Forensic Science International Vol. 287; pp. 25 - 36
Main Author: Hilal, Alaa
Format: Journal Article
Published: Elsevier B.V. Jun2018
Online Access:View this record in EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=129402446&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 129402446
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        03790738
        3L0
      jtl: Forensic Science International
      issn: 03790738
      maglogo: N
    pubinfo:
      dt: Jun2018
      vid: 287
      pid: 82545
      pub: Elsevier B.V.
      place: Philadelphia, Pennsylvania
    artinfo:
      ui:
        129402446
        129402446
        NLM29626839
        10.1016/j.forsciint.2018.03.024
        NLM29626839
        129402446
      ppf: 25
      ppct: 11
      formats:
      tig:
        atl: Image re-sampling detection through a novel interpolation kernel.
      aug:
        au: Hilal, Alaa
        affil: Department of Communication and Computer Network Engineering, Lebanese University, Aabey, Lebanon
      sug:
      ab: Image re-sampling involved in re-size and rotation transformations is an essential element block in a typical digital image alteration. Fortunately, traces left from such processes are detectable, proving that the image has gone a re-sampling transformation. Within this context, we present in this paper two original contributions. First, we propose a new re-sampling interpolation kernel. It depends on five independent parameters that controls its amplitude, angular frequency, standard deviation, and duration. Then, we demonstrate its capacity to imitate the same behavior of the most frequent interpolation kernels used in digital image re-sampling applications. Secondly, the proposed model is used to characterize and detect the correlation coefficients involved in re-sampling transformations. The involved process includes a minimization of an error function using the gradient method. The proposed method is assessed over a large database of 11,000 re-sampled images. Additionally, it is implemented within an algorithm in order to assess images that had undergone complex transformations. Obtained results demonstrate better performance and reduced processing time when compared to a reference method validating the suitability of the proposed approaches.
      pubtype: Academic Journal
      doctype: Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N