Abstract:

Shoulder surfing continues to be a serious privacy threat.  Despite this, practical and efficient countermeasures against such attacks are still scarce. We are proposing a Controlar-Freeze as an original yet effective precaution against various types of shoulder surfing attacks in ATMs in Financial Technology (FinTech). Our proposal consists of a face detection algorithm, which (a) detects if two or more people are in the scope of the camera; (b) shows an alert; (c) freezes the controls of the screen until the threat source is gone; and (d) captures the threat to be referred to as evidence. We implemented this approach on MatLab and Simulink Software. We then   conducted preliminary evaluations to validate its performance and effectiveness. Controlar-Freeze is proven success for the proposed theory that included the studied cases of the common features. We reported few concerns about this this approach as well as suggestions for improvements.