Real-time Sensor Access Security Monitor Using Windows Registry Analysis

Abstract

Think of a quiet guard always watching the computer\'s eyes and ears, the camera and microphone. This project builds exactly that: a live monitoring system for Windows that keeps constant track of which programs try to use those sensors. It works by digging into the computer\'s own registry, the digital logbook where access attempts get recorded. A small widget on the desktop instantly changes color, flashing a silent warning. At the same moment, a WhatsApp message shoots straight to the user\'s phone. To avoid constant alarms for normal use, trusted apps like Zoom or Skype go on a friendly list, a whitelist that tells the system to stand down. Need to turn the guard off from another room? A simple, lightweight web server runs locally, letting anyone start or stop the whole thing from a phone or another computer\'s browser. In a world where digital privacy feels thinner every day, this tool offers a straightforward, ever-present watch over the most personal hardware. It runs quietly in the background, barely touching the computer\'s performance, a persistent shield against spyware and unwanted snooping.

Introduction

Modern laptops contain webcams and microphones that are essential for communication but also pose a privacy risk, as malicious programs can secretly activate them without being detected by traditional antivirus software. This creates a security gap where users may be recorded or monitored without their knowledge. To address this issue, the project proposes a lightweight monitoring system for Windows that detects unauthorized access to these sensors and immediately alerts the user.

The system works by continuously monitoring the Windows Registry, which records when applications access hardware such as the camera or microphone. Whenever a program attempts to use these sensors, the system checks whether the application is trusted. If the application is not recognized, the system triggers two alerts:

• A visual indicator on the computer screen changes color (green to red).

• A WhatsApp notification is sent directly to the user’s phone.

Trusted applications like Zoom or Skype can be added to a whitelist, preventing unnecessary alerts during legitimate use. Users can also control the monitoring system remotely through a local web dashboard, allowing them to start or pause monitoring from another device on the same network.

Existing solutions for webcam security have several limitations. Basic system settings only allow users to disable sensors entirely without providing alerts or logs. Antivirus software may detect malware but does not actively monitor live camera access. Dedicated monitoring tools exist but often provide alerts only on the laptop screen, require subscriptions, or consume significant system resources. As a result, many tools function more like post-incident recorders rather than real-time protection systems.

The proposed system improves upon these limitations by offering real-time monitoring, remote alerts, and simple user control. It checks specific registry entries every second to detect sensor usage, identifies the responsible application, and compares it with a whitelist. If the application is untrusted, the system immediately triggers the alert sequence and records the event in a local log file. All data is stored locally, ensuring user privacy with no cloud storage or external data collection, except for sending a WhatsApp alert.

Testing was performed using controlled scenarios such as allowing Google Chrome to access the camera during video calls. The system successfully detected the access attempt, changed the desktop indicator color, and delivered a WhatsApp alert to the user’s phone. Trusted applications produced only a calm visual indicator without sending phone notifications.

Performance evaluation showed that the system is highly efficient and lightweight, using less than 0.5% CPU and about 30 MB of RAM during continuous operation. This makes it suitable for long-term use without affecting system performance.

Overall, the project provides a simple, practical, and privacy-focused solution that turns hidden webcam and microphone activity into visible and actionable alerts, giving users greater control over their digital privacy.

Conclusion

This project shows that feeling secure in a digital world does not require complicated locks. It needs a clear window. The tool built here acts as that window, turning the invisible activity of a computer\'s camera and microphone into a colored light you can see and a message you can hold. It works by being quiet, patient, and relentlessly observant in the background. When something unusual happens, it does not whisper a technical log; it shouts in a language you understand, right where you\'ll notice. In the end, the greatest privacy is not about stopping every possible breach. It\'s about ending the quiet fear that you might be watched without ever knowing. This work successfully trades that fear for a simple, reassuring signal.

References

[1] P. Chinnasamy, K. S. Sathya, B. J. A. Jebamani, A. Nithyasri, and S. Fowjiya, “Deep learning: Algorithms, techniques, and applications, A systematic survey,” in Deep Learning Research Applications for Natural Language Processing, IGI Global, 2023, pp. 1–17. [2] P. K. Shukla et al., “Efficient detection of unauthorized system access using registry monitoring,” IEEE Transactions on Dependable and Secure Computing, vol. 19, no. 4, pp. 211–216, 2022. [3] X. Han, R. Xie, X. Li, and J. Li, “RegMonitor: Real-time Windows Registry analysis for security threat detection,” Computers & Security, vol. 112, 2022. [4] R. Ferdousi, R. Safdari, and Y. Omidi, “Computational detection of privacy violations based on system telemetry analysis,” Journal of Cybersecurity, vol. 8, no. 1, pp. 54–64, 2021. [5] S. Liu et al., “Enhancing endpoint security using deep attention neural networks for behavioral analysis,” IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 20, no. 2, pp. 976–985, 2022. [6] X. Lin, Z. Quan, Z. Wang, T. Ma, and X. Zeng, “SecKG: Security knowledge graph for threat intelligence and attack prediction,” in Proc. 29th Int. Joint Conf. Artificial Intelligence (IJCAI), 2020, pp. 2739–2745. [7] L. H. Dang et al., “Machine learning-based prediction of unauthorized hardware access using hybrid system features,” Journal of Information Security and Applications, vol. 58, 2021. [8] P. Chinnasamy et al., “E-Security services using artificial intelligence techniques,” in Proc. Int. Conf. Computer Communication and Informatics (ICCCI), India, 2023, pp. 1–7. [9] S. Mei and K. Zhang, “A machine learning framework for predicting privacy violations in endpoint devices,” Scientific Reports, vol. 11, no. 1, p. 17619, 2021. [10] R. Chen, X. Liu, S. Jin, J. Lin, and J. Liu, “Machine learning for security threat detection and classification,” Cybersecurity, vol. 3, no. 9, 2020. [11] H. N. Ravuvar, H. Goda, S. R. and P. Chinnasamy, “Smart security monitoring system using clustering algorithms,” Proc. ICCCI, Coimbatore, India, 2020, pp. 1–4. [12] K. S. Prasad, S. Pasupathy, P. Chinnasamy and A. Kalaiarasi, “An approach to detect malicious processes using CNN-VGG16 model,” Proc. ICCCI, India, 2022, pp. 1–5. [13] P. Chinnasamy et al., “Security threat sentiment analysis using public opinions on social media,” Materials Today: Proceedings, vol. 64, pt. 1, pp. 448–451, 2022. [14] P. Chinnasamy et al., “BDDoS: Blocking distributed denial of service flooding attacks with dynamic path detectors,” Proc. ICCCI, India, 2023, pp. 1–5. [15] E. Anupriya et al., “Malicious application detection on Windows systems using machine learning techniques,” Proc. Int. Conf. Advancements in Smart, Secure and Intelligent Computing, India, 2022, pp. 1–4. [16] P. Chinnasamy et al., “An efficient phishing attack detection using machine learning algorithms,” Proc. Int. Conf. Advancements in Smart, Secure and Intelligent Computing, India, 2022, pp. 1–6. [17] P. Chinnasamy, D. Rojaramani, V. Praveena, A. J. SV, and B. Bensujin, “Data security and privacy requirements in edge computing: A systemic review,” in Cases on Edge Computing and Analytics, 2021, pp. 171–187. [18] D. Saranya et al., “Adaptive security system based on real-time monitoring for endpoint protection,” Proc. ICCCI, India, 2022, pp. 1–7. [19] P. Chinnasamy et al., “Security recommendation system using deep learning-based collaborative filtering,” Heliyon, vol. 9, no. 12, e22844, 2023. [20] P. Chinnasamy et al., “Machine learning-based cybersecurity threat prediction,” Materials Today: Proceedings, vol. 64, pt. 1, pp. 459–463, 2022. [21] T. S. Arulananth et al., “Classification of security threats using modified DenseNet-121 deep-learning model,” IEEE Access, vol. 12, pp. 35716–35727, 2024.

Copyright

Copyright © 2026 Gopi Vignesh, Kiruba M, Lakshana K, Hirthiha V. 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.