Digital Image Anti-Forensic Model Using Exponential Chaotic Biogeography-Based Optimization Algorithm

Abstract The innovation in visual imagery has led to massive growth in technologies, wherein digital cameras are obtainable at affordable prices. Thus, the digital images are easily captured and processed due to the internet connectivity. On the other hand, the development of strong image editing so...

Full description

Saved in:

Bibliographic Details
Published in:	Computer journal Vol. 66; no. 12; pp. 3038 - 3051
Main Authors:	Sudhakar, Dr R, Rao, Dr P V Venkateswara
Format:	Journal Article
Language:	English
Published:	Oxford University Press 14-12-2023
Subjects:	anti-forensics digital images JPEG compression support vector machine (SVM) artefact removal
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Abstract The innovation in visual imagery has led to massive growth in technologies, wherein digital cameras are obtainable at affordable prices. Thus, the digital images are easily captured and processed due to the internet connectivity. On the other hand, the development of strong image editing software facilitated the forgers to manipulate the accessible images with different tampering operations. Several techniques are devised for detecting the forgeries. Accordingly, this paper devises an anti-forensic model, namely Exponentially Weighted Moving Average-Chaotic Biography Based Optimization (E-CBBO) for joint photographic experts group (JPEG) compression to mitigate the forgeries occurred on the internet while transmitting data. The proposed E-CBBO is designed by integrating the properties of the exponentially weighted moving average (EWMA) with the chaotic biography-based optimization (CBBO). The suggested JPEG anti-forensic model is used to eliminate JPEG compression artifacts through the use of unique deblocking, smoothing with dither and decalibration operations. In addition, the goal is to balance visual quality and forensic undetectability when compressing the JPEG image. The fitness function is developed using the structural similarity index (SSIM), universal image quality index (UIQI) and histogram deviation parameters. With a maximum accuracy of 93.2%, a minimal MSE of 0.110, a maximum SSIM of 0.932 and a maximum UIQI of 0.890, the suggested E-CBBO beat existing approaches.
ISSN:	0010-4620 1460-2067
DOI:	10.1093/comjnl/bxac148