PlateSegFL: A Privacy-Preserving License Plate Detection Using Federated Segmentation Learning

TL;DR

The paper addresses privacy and robustness challenges in license plate detection by combining U-Net-based segmentation with Federated Learning, enabling distributed training without sharing raw data. It integrates OCR (English and Bangla) to extract plate text from segmented outputs, aiming to overcome irregular masking and improve real-time detection. The approach is validated on a multi-source dataset augmented to 11,472 images, and quantitative results show Federated U-Net outperforms bounding-box methods like YOLO in segmentation accuracy, while preserving data privacy through secure aggregation. The work demonstrates the feasibility and practical impact of privacy-preserving, segmentation-based ALPR, with potential applications in secure transportation and crime surveillance, and outlines future data expansion and digital-image deployment.

Abstract

Automatic License Plate Recognition (ALPR) is an integral component of an intelligent transport system with extensive applications in secure transportation, vehicle-to-vehicle communication, stolen vehicles detection, traffic violations, and traffic flow management. The existing license plate detection system focuses on one-shot learners or pre-trained models that operate with a geometric bounding box, limiting the model's performance. Furthermore, continuous video data streams uploaded to the central server result in network and complexity issues. To combat this, PlateSegFL was introduced, which implements U-Net-based segmentation along with Federated Learning (FL). U-Net is well-suited for multi-class image segmentation tasks because it can analyze a large number of classes and generate a pixel-level segmentation map for each class. Federated Learning is used to reduce the quantity of data required while safeguarding the user's privacy. Different computing platforms, such as mobile phones, are able to collaborate on the development of a standard prediction model where it makes efficient use of one's time; incorporates more diverse data; delivers projections in real-time; and requires no physical effort from the user; resulting around 95% F1 score.

Figures (8)

• Figure 1: System Overview of Proposed Approach Using FL U-Net Model
• Figure 2: Bangladeshi Vehicle Images
• Figure 3: Proposed U-Net model
• Figure 4: Proposed FL Architecture
• Figure 5: Extracted Vehicle Number Using OCR From FL U-Net Model Output