Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Attire-Based Anomaly Detection in Restricted Areas Using YOLOv8 for Enhanced CCTV Security

Abdul Aziz A. B, Aindri Bajpai

TL;DR

This work tackles the challenge of detecting unauthorized individuals in restricted areas by leveraging attire as a distinct biometric cue. It proposes a YOLOv8-based framework augmented with soft computing (fuzzy logic) to adapt to lighting and environmental dynamics, enabling robust, attire-based anomaly detection in CCTV footage. The methodology encompasses curated uniform-pattern datasets, transfer-learning–driven YOLOv8 training, data augmentation, and a two-stage solution architecture for person and dress detection, with adaptive thresholding and edge-friendly deployment. The results indicate fast, edge-deployable performance (approximately 8 ms per image) and improved precision through clothing-based identification, potentially reducing false alarms and enhancing security in sensitive facilities.

Abstract

This research introduces an innovative security enhancement approach, employing advanced image analysis and soft computing. The focus is on an intelligent surveillance system that detects unauthorized individuals in restricted areas by analyzing attire. Traditional security measures face challenges in monitoring unauthorized access. Leveraging YOLOv8, an advanced object detection algorithm, our system identifies authorized personnel based on their attire in CCTV footage. The methodology involves training the YOLOv8 model on a comprehensive dataset of uniform patterns, ensuring precise recognition in specific regions. Soft computing techniques enhance adaptability to dynamic environments and varying lighting conditions. This research contributes to image analysis and soft computing, providing a sophisticated security solution. Emphasizing uniform-based anomaly detection, it establishes a foundation for robust security systems in restricted areas. The outcomes highlight the potential of YOLOv8-based surveillance in ensuring safety in sensitive locations.

Attire-Based Anomaly Detection in Restricted Areas Using YOLOv8 for Enhanced CCTV Security

TL;DR

This work tackles the challenge of detecting unauthorized individuals in restricted areas by leveraging attire as a distinct biometric cue. It proposes a YOLOv8-based framework augmented with soft computing (fuzzy logic) to adapt to lighting and environmental dynamics, enabling robust, attire-based anomaly detection in CCTV footage. The methodology encompasses curated uniform-pattern datasets, transfer-learning–driven YOLOv8 training, data augmentation, and a two-stage solution architecture for person and dress detection, with adaptive thresholding and edge-friendly deployment. The results indicate fast, edge-deployable performance (approximately 8 ms per image) and improved precision through clothing-based identification, potentially reducing false alarms and enhancing security in sensitive facilities.

Abstract

This research introduces an innovative security enhancement approach, employing advanced image analysis and soft computing. The focus is on an intelligent surveillance system that detects unauthorized individuals in restricted areas by analyzing attire. Traditional security measures face challenges in monitoring unauthorized access. Leveraging YOLOv8, an advanced object detection algorithm, our system identifies authorized personnel based on their attire in CCTV footage. The methodology involves training the YOLOv8 model on a comprehensive dataset of uniform patterns, ensuring precise recognition in specific regions. Soft computing techniques enhance adaptability to dynamic environments and varying lighting conditions. This research contributes to image analysis and soft computing, providing a sophisticated security solution. Emphasizing uniform-based anomaly detection, it establishes a foundation for robust security systems in restricted areas. The outcomes highlight the potential of YOLOv8-based surveillance in ensuring safety in sensitive locations.
Paper Structure (24 sections, 5 equations, 6 figures, 3 tables, 4 algorithms)

This paper contains 24 sections, 5 equations, 6 figures, 3 tables, 4 algorithms.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. Data Collection and Preprocessing
  5. YOLOv8 Model Training
  6. Soft Computing Integration
  7. Validation and Testing
  8. System Integration and Deployment
  9. Solution Architecture
  10. Person Detection
  11. Dress Detection
  12. Uniform Detection Using YOLO
  13. Data Augmentation
  14. Network Training
  15. Observations
  16. ...and 9 more sections

Figures (6)

  • Figure 1: Simplified Workflow Architecture
  • Figure 2: Result of the Person Detector
  • Figure 3: Detecting Designation of the Person with the Attire
  • Figure 4: Training loss of the network
  • Figure 5: Comparing advanced and efficient object detectors, this overview provides key information on both response time (latency) and processing speed (throughput) at a batch size of 32. All models were tested using TensorRT 7, except for the quantized model, which utilized TensorRT 8 for evaluation.
  • ...and 1 more figures