Forensic Analysis of Colorized Grayscale Images Using Local Binary Pattern

Versatile existence of the digital images creates the need of its forensic analysis. Basically, the goal of forensic analysis is to find the malicious manipulations in the images. However, it can be divided into many categories like splicing and copy move. Sometimes various operations like contrast enhancement, histogram equalization, resampling, etc. need to be detected in forensic analysis. Also there is need of forensic analysis to distinguish between computer generated images and natural images. Now a day's various techniques are available to colorize the grayscale images. The purpose of these techniques is to color the historic grayscale images. However, to decide the credibility of images it is necessary to differentiate between natural color and colorized images. In this paper, we have applied Local Binary Pattern feature extractor method on colorized images and natural images. We have found good results on two techniques of image colorization.