Copy-move Forgery Detection Using GLCM-Based Rotation-Invariant Feature: A Preliminary Research

In the present day, the credibility of digital media is in doubt. With emerging of easy-to-use media altering software, the realistic forged media can be easily created. In the court of law, however, every single piece of digital evidence is needed to be verified. Without experts in this field, performing the basic investigation to determine the integrity of a target digital evidence is extremely hard or close to impossible. In this paper, we present a new technique for detection copy-move forgery (CMF) in the digital image using SURF key-point detection together with GLCM-based feature descriptors. Using the proposed mechanism, CMF can be efficiently detected. The experiments show detection results in which CMF area are clearly marked and shown. The proposed method not only provide support to experts or skilled practitioners but can also be used to apply with classification method (e.g. SVM) to help non-experts with automatic detection systems which need less or no human interaction.