Effective and scalable repackaged application detection based on user interface

Abstract

The app repackaging has became one of the most serious problems on Android platform nowadays. In the past, most of studies focus on the code detection, which deals with the billion opcodes, thus time consumption is expensive. Besides, it lacks the resistance to code obfuscation and hardening technology thus accuracy will easily be affected. Recently, UI-based detection was proposed. Some dynamic UI-based approaches have shown their effectiveness on repackaging detection, while all of them need to execute each app to obtain features, so it's not scalable for large apps repository analysis. In this paper, we proposed an efficient and simple approach to scalable detection, which is based on static UI features and includes two phase detection. In the coarse-grained phase, we proposed an effective dynamic selection algorithm, which is more accurate and efficient than the fixed threshold. In the fine-grained phase, we detect repackaged apps using entropy. Experimental results show that our proposed approach has a better performance on repackaged apps detection.