Computationally Efficient Digital Image Forensic Method for Image Authentication

Abstract

The issue of the authenticity and integrity of digital images is getting critical. Nowadays it became easy to create image forgeries. Digital image forensics plays a vital role in proving authenticity and integrity of digital images. There are various types of image forgeries possible, and Copy-Paste is one of it. In this type of forgery, a region of an image is copied and afterward pasted to another part of the same picture. In this paper, we propose an effective and computationally efficient method for the detection of Copy-Paste forgery. The proposed forgery detection is based on a customized Normalized Cross Correlation (NCC). The experimental results show that the proposed approach can be effectively used to detect forgeries accurately and is robust to affine transform to a certain extent.