PeekShield - Advanced Real Time Detection & Prevention System to Safeguard Against Unauthorized Screen Snooping & Visual Espionage

Abstract

The common computer settings like personal computers and laptops are some of the sources where the control of access to age-inappropriate and restricted internet content has continued to challenge control especially among children and adolescents. The current parental control systems are based on predetermined rules, manual settings, or user verification, which narrows down their application in shared device cases and allows bypass, highlighting the importance of automated age-aware access control systems [1], [3]. The article presents a new version, PeekShield, an intelligent web filtering and screen protection framework, that combines artificial intelligence, computer vision, and proxy-based viewpoint system to provide adaptive control of content. The system uses real time detection and age estimations of faces to identify minor and adult users without specifics of logging in, using previously studied age-sensitive access control and biometric authentication [1], [3].A local age-checking service is dynamically applied to implement browsing policies through a man-in-the-middle proxy design, and is aligned with traditional secure web filtering designs [4], [6].Also, PeekShield has visual privacy by identifying and blocking unknown faces around the screen and automatically blurring screen content or locking down to stop shoulder surfing, as confirmed by screen privacy research [2], [5].In general, PeekShield offers a privacy-sensitive and scalable solution to the securement of shared computing environments.

Introduction

The document presents PeekShield, an AI-based security system designed to protect users in shared computing environments such as homes, schools, and public systems. It addresses key limitations of traditional parental control and web filtering tools, which rely on static rules and user authentication and are often ineffective in shared-device scenarios.

PeekShield introduces a dynamic, context-aware approach using computer vision and artificial intelligence. It continuously captures webcam input to detect users and estimate their age, classifying them as minors or adults. Based on this classification, the system automatically applies appropriate access control policies without requiring manual login or user accounts.

To enforce safety, PeekShield uses a proxy-based network filtering system that blocks inappropriate or age-restricted content at the network level before it reaches the browser. In addition to digital protection, it also ensures physical privacy by monitoring the surrounding environment through the webcam. If unauthorized or unknown faces are detected, the system can blur or lock the screen to prevent shoulder-surfing and visual data exposure.

The system includes a monitoring dashboard that provides real-time updates on user activity, blocked content, and security events for administrators or guardians.

Experimental results show that PeekShield performs effectively, achieving high face detection accuracy (around 97%), fast intrusion response (under 1 second), and over 94% accuracy in age-based classification. It also supports multi-user scenarios, real-time logging, remote access requests, and efficient system performance with low resource usage.

Conclusion

This work introduced PeekShield, a context-sensitive and intelligent protection system with an approach of dealing with the increased privacy and security problems with shared computing devices. In contrast to the previous parental control systems and content filtering systems, which utilize a fixed set of rules, configurations, or hard-coded passwords, PeekShield presents a new system that is automatically adjusted to the user environment and active user, adapting in line with it. The system has a combination of computer vision, access control based on age and network level enforcement that gives the system all over protection against exposure of the digital and physical observation threats. PeekShield is an implementation that integrates live face recognition and age estimation with a policy engine that can be run locally to make access control decisions as they happen and address privacy concerns. The adoption of proxy-based enforcement mechanism will protect the blocking of restricted content such that it is blocked at network layer before it reaches an interface making it more robust and less vulnerable to typical bypass mechanisms. Moreover, that the system can spot intruders as well as automatically enforce protection that obstructs the view of the screen combats a very important but rarely considered area of shared-device security. The presence of a mechanism of monitoring and visualization also advantages the usefulness of the PeekShield by increasing the levels of transparency and confidence in the system. The framework helps to reduce the broad gap between intelligent automation and human control by displaying system actions and decisions clearly and, therefore, the solution can be used successfully in long-term systems in real-world settings, as homes, educational institutions, and public access systems. By and large, PeekShield brings to the fore age-conscious content filtering, visual privacy protection, and enforcement of cybersecurity in a single unified architecture. The presented system proves that it is possible to produce adaptive, automated, and user oriented security solutions that can autonomously handle the complicated issues of modern shared computing settings and retain usability, privacy and accountability. This collective strategy will provide a solid foundation of intelligent security systems in the future that will have to work safely and openly in more shared and dynamic online settings.

References

[1] Ranjan, R., Patel, S., & Kumar, A. (2020). Age estimation using deep convolutional neural networks for access control systems. IEEE Access, 8, 112345–112356. [1] [2] Park, J., & Kim, H. (2020). Privacy-preserving screen protection against shoulder surfing using computer vision. Computers & Security, 95, 101–112.[2] [3] Li, S. Z., Lei, Z., & Yang, M. (2019). Face detection and age estimation in unconstrained environments. Pattern Recognition Letters, 124, 36–42. [3] [4] Provos, N., & McNamee, D. (2018). Proxy-based content filtering for secure web access. ACM Transactions on Information and System Security, 14(2), 1–24.[4] [5] Kapadia, A., Marforio, C., & Nguyen, H. (2019). Mitigating shoulder surfing attacks through adaptive screen privacy. IEEE Security & Privacy, 17(3), 42–50. [5] [6] Mitmproxy Developers. (2023). Mitmproxy: A free and open-source interactive HTTPS proxy. [6] [7] Livingstone, S., & Helsper, E. J. (2010). Balancing opportunities and risks in teenagers’ use of the internet: The role of online skills and internet self-efficacy. New Media & Society, 12(2), 309–329.[7] [8] Nguyen, Park, J., & Kim, H. (2020). Privacy-preserving screen protection against shoulder surfing using computer vision. Computers & Security, 95, 101–112. [8] [9] Wisniewski, P., Xu, H., Rosson, M. B., & Carroll, J. M. (2017). Parents just don’t understand: Why teens don’t talk to parents about their online risk experiences. Proceedings of the ACM Conference on Computer Supported Cooperative Work (CSCW), 523–540. [9] [10] Levi, G., & Hassner, T. (2015). Age and gender classification using convolutional neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 34–42S. [10] [11] Eiband, M., Khamis, M., von Zezschwitz, E., & Hussmann, H. (2017). Understanding shoulder surfing in the wild: Stories from users and observers. Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems, 4254–4265.[11]

Copyright

Copyright © 2026 Erukonda Mohan Manoj, Punna Roshni, Pippalapalli Purushotham, N. Mounika. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.