An Event-Driven Edge AI Surveillance System with Real-Time Automated Alerting
International Journal of Creative and Open Research in Engineering and Management · 20.11.2026 · 02 (11) · 1-9
DOI: 10.55041/ijcope.v2i11.125 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 20.11.2026. Verfasst von Akash S und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Traditional surveillance networks rely on continuous video recording, which demands immense data storage, consumes high network bandwidth, and serves primarily as a passive forensic tool rather than an active deterrent. While basic motion-activated systems mitigate storage waste, they suffer from critically high false-alarm rates triggered by environmental noise. To resolve this, we propose an event-driven Edge AI surveil-lance architecture using a Raspberry Pi 3 Model B+ paired with an OV5647 NoIR camera. The system utiliz-es a time-domain decoupled approach: an NS312 PIR sensor acts as a strict zero-power hardware interrupt, keeping the processing pipeline dormant until physical motion is detected. Upon activation, an onboard OpenCV Haar Cascade classifier executes instantaneous face verification to definitively confirm human presence, completely eliminating false positives. The Python-based firmware employs multithreaded concurrency to simul-taneously trigger a localized 5V active piezo buzzer for on-site deterrence, dispatch annotated visual alerts via the Telegram Bot API, and maintain a thread-safe Flask MJPEG web stream. Physical validation on Debian 13 "Trixie" demonstrates stable zero-light operation, miti-gation of I2C brownouts, and sub-3-second alert latency without reliance on cloud-based AI. Keywords- Edge AI, Event-Driven Architecture, OpenCV Haar Cascade, Telegram Bot API, Hardware Interrupt, Surveillance, False Alarm Mitigation..
Zitieren
APA
S, A., Badiger, S. C., C, V. K., M, S., & M, D. K. (2026). An Event-Driven Edge AI Surveillance System with Real-Time Automated Alerting. International Journal of Creative and Open Research in Engineering and Management, 02(11), 1-9. https://doi.org/10.55041/ijcope.v2i11.125
Vancouver
S A, Badiger SC, C VK, M S, M DK. An Event-Driven Edge AI Surveillance System with Real-Time Automated Alerting. International Journal of Creative and Open Research in Engineering and Management. 2026;02(11):1-9. doi:10.55041/ijcope.v2i11.125
BibTeX
@article{S2026,
author = {S, Akash and Badiger, Shrilakshmi C and C, Vamshi Krishna and M, Sathwik and M, Dr. Kishore},
title = {An Event-Driven Edge AI Surveillance System with Real-Time Automated Alerting},
journal = {International Journal of Creative and Open Research in Engineering and Management},
year = {2026},
volume = {02},
number = {11},
pages = {1--9},
doi = {10.55041/ijcope.v2i11.125}
}
EUVIMED