Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Novel AI-Driven System for Real-Time Detection of Mirror Absence, Helmet Non-Compliance, and License Plates Using YOLOv8 and OCR

Nishant Vasantkumar Hegde, Aditi Agarwal, Minal Moharir

TL;DR

This work addresses automated, real-time enforcement of two-wheeler safety regulations by detecting helmet non-compliance, rear-view mirror absence, and extracting license plates. It proposes a holistic pipeline based on YOLOv8 for multi-class detection, EasyOCR for license plate recognition, and a Streamlit-based frontend for monitoring and logging. A novel aspect is the inclusion of rear-view mirror absence as a detection target, coupled with an OCR preprocessing pipeline that improves readability under challenging conditions. On a custom Indian-road dataset, the system achieves high precision and competitive mAP, enabling scalable, real-time enforcement suitable for deployment in traffic management settings.

Abstract

Road safety is a critical global concern, with manual enforcement of helmet laws and vehicle safety standards (e.g., rear-view mirror presence) being resource-intensive and inconsistent. This paper presents an AI-powered system to automate traffic violation detection, significantly enhancing enforcement efficiency and road safety. The system leverages YOLOv8 for robust object detection and EasyOCR for license plate recognition. Trained on a custom dataset of annotated images (augmented for diversity), it identifies helmet non-compliance, the absence of rear-view mirrors on motorcycles, an innovative contribution to automated checks, and extracts vehicle registration numbers. A Streamlit-based interface facilitates real-time monitoring and violation logging. Advanced image preprocessing enhances license plate recognition, particularly under challenging conditions. Based on evaluation results, the model achieves an overall precision of 0.9147, a recall of 0.886, and a mean Average Precision (mAP@50) of 0.843. The mAP@50 95 of 0.503 further indicates strong detection capability under stricter IoU thresholds. This work demonstrates a practical and effective solution for automated traffic rule enforcement, with considerations for real-world deployment discussed.

A Novel AI-Driven System for Real-Time Detection of Mirror Absence, Helmet Non-Compliance, and License Plates Using YOLOv8 and OCR

TL;DR

This work addresses automated, real-time enforcement of two-wheeler safety regulations by detecting helmet non-compliance, rear-view mirror absence, and extracting license plates. It proposes a holistic pipeline based on YOLOv8 for multi-class detection, EasyOCR for license plate recognition, and a Streamlit-based frontend for monitoring and logging. A novel aspect is the inclusion of rear-view mirror absence as a detection target, coupled with an OCR preprocessing pipeline that improves readability under challenging conditions. On a custom Indian-road dataset, the system achieves high precision and competitive mAP, enabling scalable, real-time enforcement suitable for deployment in traffic management settings.

Abstract

Road safety is a critical global concern, with manual enforcement of helmet laws and vehicle safety standards (e.g., rear-view mirror presence) being resource-intensive and inconsistent. This paper presents an AI-powered system to automate traffic violation detection, significantly enhancing enforcement efficiency and road safety. The system leverages YOLOv8 for robust object detection and EasyOCR for license plate recognition. Trained on a custom dataset of annotated images (augmented for diversity), it identifies helmet non-compliance, the absence of rear-view mirrors on motorcycles, an innovative contribution to automated checks, and extracts vehicle registration numbers. A Streamlit-based interface facilitates real-time monitoring and violation logging. Advanced image preprocessing enhances license plate recognition, particularly under challenging conditions. Based on evaluation results, the model achieves an overall precision of 0.9147, a recall of 0.886, and a mean Average Precision (mAP@50) of 0.843. The mAP@50 95 of 0.503 further indicates strong detection capability under stricter IoU thresholds. This work demonstrates a practical and effective solution for automated traffic rule enforcement, with considerations for real-world deployment discussed.

Paper Structure

This paper contains 18 sections, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background Study and Related Work
  3. Methodology
  4. Model Selection and Object Detection
  5. Dataset Creation and Annotation
  6. Image Preprocessing for OCR
  7. Violation Detection Logic
  8. License Plate Recognition (LPR)
  9. Frontend Development and Data Logging
  10. Results and Discussion
  11. Overall Detection Performance
  12. Model Training Dynamics and Analysis
  13. Qualitative Analysis and OCR Performance
  14. Ablation Study on OCR Preprocessing
  15. Real-time Processing and Comparative Analysis
  16. ...and 3 more sections

Figures (4)

  • Figure 1: Block Diagram of the Proposed AI-Driven Traffic Violation Detection System. This illustrates the sequential flow from input processing and object detection with YOLOv8, through OCR-based license plate recognition, to violation logging and display via the Streamlit interface.
  • Figure 2: Model Training Dynamics: (a) Training Loss. (b) Validation Loss. (c) P, R, & mAP. (d) Learning Rate. (e) F1-Confidence. (f) Confusion Matrix.
  • Figure 3: System output examples: (a) Multiple violators. (b) Rider/pillion non-compliance. (c) LP detection & OCR. (d) No violations.
  • Figure 4: Streamlit interface violation log.