Cyber Forensic Security On Multimedia Communication

Abstract

ABSTRACT newline newline newline newline newlineForensic security for digital multimedia data has been broadly recognized as a capable model to defeat the cyber crime activities during digital data transfer. Financial documents, medical records, journalism and reports are some real-world applications which use digital data as input. Forensic security on digital multimedia image is attained by using spatial resolution forensic image segmentation mechanism. There are various technologies that are designed for securing the multimedia data information by occupying enormous time of security. Hence, proposed research work focuses on improving the forensic security on digital multimedia data. newlineExisting Xbox One restores Forensic images were developed to remove the hard drive from the system utilizing hardware write blocker. They provide security on various applications when the images are encrypted. Fingerprint matching using graphic processing improves fingerprint matching. Additionally, it save the video game log file on the hard drive. However, adding security on digital features caused huge complexity in forensically acquiring digital forensic artifacts. newlinePhoto-Response Non-Uniformity (PRNU) technique uses modern convex optimization based approach to discover image forgeries with the help of detection algorithms in the image forgery system. Here, Markov random field is used to relocate the dependence of source and make decision for each pixel images. Nonlocal denoising is routed to improve the modern convex optimization method. However, less significant forgeries with respect to spatial resolution remained unaddressed. newlineFingerprint Matching using Graphics Processing Units (FM-GPU) provides matching model known as minutiae matching. The minutiae matching on fingerprint is simple and similarly it does not achieve higher forensic accuracy rate. In addition, content-based fingerprint matching fails to offer the robustness against content change attacks. Another existing named as Sparse Reconstruction based Metric Learning Method (SRMLM) uses the po