What question did this study set out to answer?

Develop a robust, real-time automatic license plate recognition (ALPR) system capable of maintaining high accuracy across diverse vehicle types and unconstrained operating conditions.

July 1, 2018Open Access

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

View Full Paper

Ask AI

Bookmark

Authors

RLRayson LarocaPontifícia Universidade Católica do Paraná ESEvair Borges SeveroHôpital intercantonal de la Broye LZLuiz A. ZanlorensiUniversidade Tecnológica Federal do Paraná

Discussion

Member takes

Overview

Benchmark evaluation demonstrates superior real-time license plate recognition across moving vehicle videos, highlighting improved robustness over commercial systems.

Key Points

Develop a robust, real-time automatic license plate recognition (ALPR) system capable of maintaining high accuracy across diverse vehicle types and unconstrained operating conditions.
Trained and fine-tuned convolutional neural networks based on the YOLO detector across each ALPR stage, incorporating a two-stage character segmentation and recognition pipeline.
Implemented data augmentation strategies, including plate inversion and character flipping, to enhance tolerance to lighting, camera, and background variations.
Evaluated system accuracy and throughput on the SSIG benchmark (2,000 frames from 101 videos) and a newly introduced UFPR-ALPR dataset (4,500 frames from 150 videos) recorded under dynamic camera and vehicle motion.
Achieved a 93.53% recognition rate at 47 frames per second (FPS) on the SSIG dataset, surpassing Sighthound (89.80%), OpenALPR (93.03%), and the previous benchmark (81.80%).
Attained a 78.33% recognition rate at 35 FPS on the UFPR-ALPR dataset, whereas trial versions of commercial ALPR systems scored below 70%.

Cite This Study

Laroca et al. (2018) studied this question.

synapsesocial.com/papers/6a00a9847ac91c5d2a2d6e10 https://doi.org/10.1109/ijcnn.2018.8489629

View Full Paper

Ask AI

Bookmark

July 1, 2018Open Access

A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector

View Full Paper

Ask AI

Bookmark

Authors

RLRayson LarocaPontifícia Universidade Católica do Paraná ESEvair Borges SeveroHôpital intercantonal de la Broye LZLuiz A. ZanlorensiUniversidade Tecnológica Federal do Paraná

Discussion

Member takes

Overview

Benchmark evaluation demonstrates superior real-time license plate recognition across moving vehicle videos, highlighting improved robustness over commercial systems.

Key Points

Develop a robust, real-time automatic license plate recognition (ALPR) system capable of maintaining high accuracy across diverse vehicle types and unconstrained operating conditions.
Trained and fine-tuned convolutional neural networks based on the YOLO detector across each ALPR stage, incorporating a two-stage character segmentation and recognition pipeline.
Implemented data augmentation strategies, including plate inversion and character flipping, to enhance tolerance to lighting, camera, and background variations.
Evaluated system accuracy and throughput on the SSIG benchmark (2,000 frames from 101 videos) and a newly introduced UFPR-ALPR dataset (4,500 frames from 150 videos) recorded under dynamic camera and vehicle motion.
Achieved a 93.53% recognition rate at 47 frames per second (FPS) on the SSIG dataset, surpassing Sighthound (89.80%), OpenALPR (93.03%), and the previous benchmark (81.80%).
Attained a 78.33% recognition rate at 35 FPS on the UFPR-ALPR dataset, whereas trial versions of commercial ALPR systems scored below 70%.

Cite This Study

Laroca et al. (2018) studied this question.

synapsesocial.com/papers/6a00a9847ac91c5d2a2d6e10 https://doi.org/10.1109/ijcnn.2018.8489629

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark