On the Cross-dataset Generalization in License Plate Recognition

Abstract

Automatic License Plate Recognition (ALPR) systems have shown remarkable performance on license plates (LPs) from multiple regions due to advances in deep learning and the increasing availability of datasets. The evaluation of deep ALPR systems is usually done within each dataset; therefore, it is questionable if such results are a reliable indicator of generalization ability. In this paper, we propose a traditional-split versus leave-one-dataset-out experimental setup to empirically assess the cross-dataset generalization of 12 Optical Character Recognition (OCR) models applied to LP recognition on nine publicly available datasets with a great variety in several aspects (e.g., acquisition settings, image resolution, and LP layouts). We also introduce a public dataset for end-to-end ALPR that is the first to contain images of vehicles with Mercosur LPs and the one with the highest number of motorcycle images. The experimental results shed light on the limitations of the traditional-split protocol for evaluating approaches in the ALPR context, as there are significant drops in performance for most datasets when training and testing the models in a leave-one-dataset-out fashion.

Accepted for presentation at the International Conference on Computer Vision Theory and Applications (VISAPP) 2022

References in corpus (3)

• YOLOv4: Optimal Speed and Accuracy of Object Detection
• Convolutional Neural Networks for Automatic Meter Reading
• Tell me what you see: A zero-shot action recognition method based on natural language descriptions

Cited by in corpus (13)

• Super-Resolution of License Plate Images Using Attention Modules and Sub-Pixel Convolution Layers
• A First Look at Dataset Bias in License Plate Recognition
• Combining Attention Module and Pixel Shuffle for License Plate Super-Resolution
• Do We Train on Test Data? The Impact of Near-Duplicates on License Plate Recognition
• Leveraging Model Fusion for Improved License Plate Recognition
• Toward Advancing License Plate Super-Resolution in Real-World Scenarios: A Dataset and Benchmark
• Enhancing License Plate Super-Resolution: A Layout-Aware and Character-Driven Approach
• Toward Enhancing Vehicle Color Recognition in Adverse Conditions: A Dataset and Benchmark
• Multi-Feature Aggregation in Diffusion Models for Enhanced Face Super-Resolution
• Advancing Multinational License Plate Recognition Through Synthetic and Real Data Fusion: A Comprehensive Evaluation
• LPLC: A Dataset for License Plate Legibility Classification
• Toward Unified Fine-Grained Vehicle Classification and Automatic License Plate Recognition
• DACov: A Deeper Analysis of Data Augmentation on the Computed Tomography Segmentation Problem