Front Inner Page - Volume 2 No.5 October 2015

  • » Back to Index

  • Title

    :

    An Efficient Copy-Move Forgery Detection scheme using Segmentation

    Authors

    :

    B.Femina 1 , T.Chelladurai 2

    Keywords

    :

    Copy-Move Forgery (CMF), Expectation Maximization (EM), Scale-Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF), Affine transform matrix, Segmentation.

    Issue Date

    :

    October – 2015

    Abstract

    :

    The development of new hardware and improved human-computer interfaces can effortlessly generate the fake images. Multiple image processing softwares are used to detect the image forgeries that could not be seen through the human eye. In the Copy-Move Forgery (CMF), a part of the image is copied and pasted in the other part of the same image. In this paper, an efficient segmentation based copy-move forgery detection scheme is proposed. When compared to the existing key point extractionbased state-of-the-art forgery detection schemes,the proposed scheme splits the images into independent patches. The copy-move forgery regions are detected by matching the image patches. The matching process consist of two stages. During the initial stage, the pairs that contain the copy-move forgery regions are detected. With the computed patches, an affine transform matrix is constructed. In the second stage, the estimated matrix is optimized using the Expectation-Maximization (EM) based algorithm. Further, the presence of the copy-move forgery regions are confirmed. The performance of the proposed scheme is compared with the existing Scale-Invariant Feature Transform (SIFT), Zemike and the Speeded-Up Robust Features (SURF) algorithms for the performance metrics such as precision and recall. The experimental results show that the proposed scheme is optimal than the existing algorithms.  

    Page(s)

    :

    31-38

    ISSN

    :

    2347- 4734

    Source

    :

    Vol. 2, No.5, October 2015

    Download

    :


  • » Back index