Indian Journal of Science and Technology

Abstract

Nowadays digital media manipulation has become a common trend. Digital media, especially images, being one of the primary modes of communication, can be easily manipulated. Current research trends in digital image forensics focus on validating the authenticity of the image. Objectives: Objective of the present study is to authenticate objects in an image using light sources and their properties. Method: By locating the direction of the light source, forgery in the images can be easily detected. Inconsistencies between different light sources in the image highlight image tampering. This paper focuses on detection of image forgery using lighting inconsistencies. The proposed technique measures the lighting properties from different objects or surfaces present in the image. Finding: The model for digital forensics identifies the lighting discrepancies in the objects of an image and provides results indicating difference between real and fake images. Improvement: The proposed technique is an objective based method and identifies digital image forgery based on physics of the environment. Results are promising and reproducible making the technique an important tool for image forgery detection.
Keywords: Authenticity, Image Forensics, Image Tampering Detection, Lighting