ApexGuard-AI: A Unified Real-Time Multi-Threat Surveillance System using YOLOv8/v9 Deep Learning Models

Amey Sanjay Avasthi; Piyush Vishal Patil; Divyaraj Vijay Patil; Jayesh Vinodrao Sonawane; Sarang Sanjay Deore

ApexGuard-AI: A Unified Real-Time Multi-Threat Surveillance System using YOLOv8/v9 Deep Learning Models

Author(s)	Mr. Amey Sanjay Avasthi, Mr. Piyush Vishal Patil, Mr. Divyaraj Vijay Patil, Mr. Jayesh Vinodrao Sonawane, Mr. Sarang Sanjay Deore
Country	India
Abstract	Traditional CCTV surveillance depends on human operators susceptible to fatigue, delayed response, and limitedscalability. This paper presents ApexGuard-AI, a unified real-time multi-threat AI surveillance platform that concurrentlydetects dangerous wildlife, fire and smoke, and weapons from a single camera feed. The system employs a custom fine-tunedYOLOv8n model for biological threat identification, a dedicated YOLOv9 model for fire and smoke recognition, and a standardYOLOv8n COCO model with label filtering for weapon detection. A FastAPI asynchronous backend manages video streaming and alert dispatch while a daemon detection thread overcomes Python's Global Interpreter Lock (GIL) limitation. Events arepersisted through a dual-stage CSV-to-SQLite logging pipeline, and a rolling 40-frame pre-event buffer ensures completeincident capture. Experimental evaluation on standard consumer hardware demonstrates 25 FPS throughput with alert latencybelow 150 ms, validating the practical feasibility of automated, multi-class AI threat detection in real-world surveillanceenvironments.
Keywords	YOLOv8, YOLOv9, Real-time surveillance, Fire detection, Wildlife detection, Weapon detection, FastAPI, Deep Learning, CCTV Automation
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-13

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

ApexGuard-AI: A Unified Real-Time Multi-Threat Surveillance System using YOLOv8/v9 Deep Learning Models

Share this