Unmasking Android Obfuscation Tools Using Spatial Analysis

Android has become one of the most popular mo-bile device operating systems. Indeed, its security issues have attracted a lot of attention. One of the major security concerns is the use of obfuscation strategies to evade anti-malware solutions. Android malware authors are increasingly using code obfuscation tools and techniques to hide malicious code. In this work, we in- troduce a novel fingerprinting approach for Android obfuscation tools based on spatial analysis. We investigate first-order and second-order statistical features to analyze spatial distribution of pixels representing Android binary images. With our approach, we are able to achieve nearly 90% accuracy in fingerprinting several obfuscation tools with specific configuration options.