Analysis designing and evaluating technique to extenuate malware spread through images over online social networking sites

Abstract

In this thesis, social networking sites have been analyzed which are used to share information through images for identifying the different types of threats associated with them. Numerous algorithms have been explored related to JPEG steganography and JPEG steganalysis techniques to understand the JPEG markers, their types and working with the detection accuracy. Real attacks which are occurred through malicious jpeg attacks are analyzed for malicious codes or scripts inside the images, which are intentionally embedded by the attackers. In this research work, a novel framework is proposed and implemented to detect the malicious code/script/URL present inside the JPEG image. A complete secured virtual malware analysis lab was developed in which the working of real malicious samples are explored and noted. For this research 300 malicious JPEG image samples are collected from five different antivirus companies repositories. Through this research the output comes as a complete way to analyze any JPEG image for detection and extenuate a malware which spread through images over online social networking sites.