UIHash: Detecting Similar Android UIs through Grid-Based Visual Appearance Representation

Watch a 12-minute conference presentation from USENIX Security '24 exploring a novel approach to detecting similar Android user interfaces through grid-based visual appearance representation. Learn about UIHash, a system developed by researchers from Beihang University and the National University of Singapore that addresses security concerns related to deceptive mobile app interfaces. Discover how the system divides UI screens into a #-shaped grid pattern and uses a neural network architecture combining CNN and Siamese networks to detect similar UIs with an impressive F1-score of 0.984. Understand the limitations of existing tree-based and image-based detection methods, and explore how UIHash's innovative approach provides better protection against UI spoofing and phishing attacks by focusing on how users actually perceive and interact with mobile interfaces.