An Image Forensic Technique Based on SIFT Descriptors and FLANN Based Matching

EasyChair Preprint 6828

Abstract

Doctored images are prevalent everywhere since the easy availability of photo editing tools. The research in the field of image forensics is focused mainly on developing techniques that can help to discriminate between a doctored and a legitimate content in an image. There are various kinds of forgeries possible in an image. Here, we propose a robust algorithm for the detection of copy-move forgery. We exploit the simple linear iterative clustering (SLIC) algorithm to divide the given image into non-overlapping, irregular-sized blocks and then use Scale Invariant Feature Transform (SIFT) to extract the feature key points. Thereafter, Fast Library for Approximate Nearest Neighbors (FLANN) is used to match the key points between blocks. Forged regions are chalked out accurately employing some morphological operations and analysis using correlation coefficient. To validate the efficacy of the proposed algorithm, we have tested it on four standard datasets and found out the proposed scheme is performing satisfactorily well. It is robust against scaling, JPEG compression, and rotation.

Keyphrases: Copy-move forgery, FLANN matching, SIFT, SLIC

@booklet{EasyChair:6828,
  author    = {Megha Gupta and Priyanka Singh},
  title     = {An Image Forensic Technique Based on SIFT Descriptors and FLANN Based Matching},
  howpublished = {EasyChair Preprint 6828},
  year      = {EasyChair, 2021}}