Toward Advancing License Plate Super-Resolution in Real-World Scenarios: A Dataset and Benchmark

2025-05-09

AI Summary

To address the severe degradation in license plate recognition (LPR) performance caused by low-resolution and degraded images in real-world surveillance scenarios, this work introduces UFPR-SR-Plates—the first large-scale, trajectory-driven real-world license plate super-resolution dataset, comprising 10,000 vehicle trajectories and 100,000 LR/HR image pairs—and establishes a multi-frame sequential super-resolution benchmark. We propose a character-level majority-voting fusion strategy (MVCP), integrated with the LCDNet super-resolution model, OCR-based post-processing, and spatiotemporal frame alignment. Experiments demonstrate that LPR accuracy improves from 1.7% on raw low-resolution inputs to 31.1% using single-frame super-resolution, and further to 44.7% using MVCP fusion over five frames. This is the first empirical validation of the critical synergistic gain achieved by jointly leveraging super-resolution and multi-frame temporal information for robust LPR in realistic settings.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Large-scale security measurements

Abstract

Recent advancements in super-resolution for License Plate Recognition (LPR) have sought to address challenges posed by low-resolution (LR) and degraded images in surveillance, traffic monitoring, and forensic applications. However, existing studies have relied on private datasets and simplistic degradation models. To address this gap, we introduce UFPR-SR-Plates, a novel dataset containing 10,000 tracks with 100,000 paired low and high-resolution license plate images captured under real-world conditions. We establish a benchmark using multiple sequential LR and high-resolution (HR) images per vehicle -- five of each -- and two state-of-the-art models for super-resolution of license plates. We also investigate three fusion strategies to evaluate how combining predictions from a leading Optical Character Recognition (OCR) model for multiple super-resolved license plates enhances overall performance. Our findings demonstrate that super-resolution significantly boosts LPR performance, with further improvements observed when applying majority vote-based fusion techniques. Specifically, the Layout-Aware and Character-Driven Network (LCDNet) model combined with the Majority Vote by Character Position (MVCP) strategy led to the highest recognition rates, increasing from 1.7% with low-resolution images to 31.1% with super-resolution, and up to 44.7% when combining OCR outputs from five super-resolved images. These findings underscore the critical role of super-resolution and temporal information in enhancing LPR accuracy under real-world, adverse conditions. The proposed dataset is publicly available to support further research and can be accessed at: https://valfride.github.io/nascimento2024toward/

Problem

• Addressing low-resolution license plate recognition challenges
• Introducing a real-world dataset for super-resolution benchmarking
• Improving OCR accuracy via fusion of super-resolved images

Innovation

• Introduces UFPR-SR-Plates dataset with real-world images
• Uses LCDNet model for license plate super-resolution
• Applies MVCP fusion strategy to boost OCR accuracy