Robustness of Color Interpolation Identification against Anti-forensic Operations

Color interpolation identification using digital images has been shown to be a powerful tool for addressing a range of digital forensic questions. However, due to the existence of adversaries who have the incentive to counter the identification, it is necessary to understand how color interpolation identification performs against anti-forensic operations that intentionally manipulate identification results. This paper proposes two anti-forensic techniques against which the robustness of color interpolation identification is investigated. The first technique employs parameter perturbation to circumvent identification. Various options that achieve different trade-offs between image quality and identification manipulation are examined. The second technique involves algorithm mixing and demonstrates that one can not only circumvent but also mislead the identification system while preserving the image quality. Additional discussions are also provided to enhance the understanding of anti-forensics and its implications to the design of identification systems.