Adaptive Bilateral Filtering Detection Using Frequency Residuals for Digital Image Forensics
A. U. Shehin, Deepa Sankar · 2022
The credibility of digital images is crucial nowadays. Digital images bear ample information, and they even stand as evidence in a court of law. But editing an image is effortless due to easily usable image editing softwares. So image editing detection has become a hot topic for researchers recently. Over time, several antiforensic techniques like blurring and median filtering have emerged to mask image forgery. While several types of research are done to detect blurring and median filtering, this paper introduces the detection of another powerful antiforensic tool called adaptive bilateral filtering. The proposed method develops a new frequency domain residual feature by successive adaptive bilateral filtering of an image. The polarity of this feature can effectively detect whether an image is previously adaptive bilateral filtered or not. The experiments performed in this work show that the proposed detection method is more effective than commonly used median filtering detection performed in the frequency domain.