On the Cross-dataset Generalization in License Plate Recognition

Laroca, Rayson; Cardoso, E.S.; Lucio, Diego Rafael; Estevam, Valter; Menotti, David

doi:10.5220/0010846800003124

Cited by 41 publications

(16 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We made use of LP images obtained from the RodoSol-ALPR [18] and PKU [19] datasets. To the best of our knowledge, there is currently no public dataset that provides paired LR and HR images from real-world settings.…”

Section: A Setupmentioning

confidence: 99%

“…3 the diversity of this dataset regarding several factors such as LP colors, lighting conditions, and character fonts. Here, we follow the standard protocol (defined in [18]) that involves using 40% of the images for training, 20% for validation, and 40% for testing. Fig.…”

Section: A Setupmentioning

confidence: 99%

“…3. Some LP images from the RodoSol-ALPR dataset [18]. The first two rows show Brazilian LPs, while the last two rows show Mercosur LPs.…”

Section: A Setupmentioning

confidence: 99%

“…In this work, we report the results of experiments performed on two datasets, collected in different regions under various conditions, instead of a single one. In the RodoSol-ALPR dataset [18], our approach recognizes at least five characters in 74.2% of LPs compared to 42.2% by our previous model trained and evaluated under the same conditions. In the PKU dataset [19], the improvement was even more significant, from 82.5% by the preliminary approach to 97.3% by the improved one (proposed in this work).…”

Section: Introductionmentioning

confidence: 95%

See 3 more Smart Citations

Super-resolution of license plate images using attention modules and sub-pixel convolution layers

Nascimento

Laroca

Lambert

et al. 2023

Computers & Graphics

View full text Add to dashboard Cite

Section: A Setupmentioning

confidence: 99%

Section: A Setupmentioning

confidence: 99%

“…3. Some LP images from the RodoSol-ALPR dataset [18]. The first two rows show Brazilian LPs, while the last two rows show Mercosur LPs.…”

Section: A Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 95%

See 2 more Smart Citations

Super-resolution of license plate images using attention modules and sub-pixel convolution layers

Nascimento

Laroca

Lambert

et al. 2023

Computers & Graphics

View full text Add to dashboard Cite

“…One of the problems pointed out in [Krinski et al 2021] was the class imbalance due to several images with just the background class; in fact, recent work has shown that several problems suffer from class imbalance [Laroca et al 2021, Laroca et al 2022. Therefore, to mitigate this problem, in the first step of this evaluation, we removed from the datasets images with no lesion in the ground-truth mask.…”

Section: Metrics and Datasetsmentioning

confidence: 99%

Light In The Black: An Evaluation of Data Augmentation Techniques for COVID-19 CT's Semantic Segmentation

Krinski¹,

Ruíz²,

Todt³

2022

Preprint

View full text Add to dashboard Cite

With the COVID-19 global pandemic, computer-assisted diagnoses of medical images have gained much attention, and robust methods of Semantic Segmentation of Computed Tomography (CT) became highly desirable. Semantic Segmentation of CT is one of many research fields of automatic detection of COVID-19 and has been widely explored since the COVID-19 outbreak. In this work, we propose an extensive analysis of how different data augmentation techniques improve the training of encoder-decoder neural networks on this problem. Twenty different data augmentation techniques were evaluated on five different datasets. Each dataset was validated through a five-fold crossvalidation strategy, thus resulting in over 3,000 experiments. Our findings show that spatial level transformations are the most promising to improve the learning of neural networks on this problem.Resumo. Com a COVID-19, diagnósticos de imagens médicas assistidos por computador ganharam muita atenc ¸ão, e métodos robustos de Segmentac ¸ão Semântica de Tomografia Computadorizada (TC) tornaram-se altamente desejáveis. A Segmentac ¸ão Semântica de TC é um dos muitos campos de pesquisa de detecc ¸ão automática da COVID-19 e foi amplamente explorado desde o surto da COVID-19. Neste trabalho, propomos uma análise extensiva sobre o quanto diferentes técnicas de aumento de dados contribuem para melhorar o treinamento de redes neurais codificador-decodificador sobre este problema. Vinte técnicas diferentes de aumento de dados foram avaliadas em cinco conjuntos de dados diferentes. Cada conjunto de dados foi validado através de uma estratégia de validac ¸ão cruzada de cinco subconjuntos, resultando assim em mais de 3.000 experimentos. Nossas descobertas mostram que as transformac ¸ões de nível espacial são as mais promissoras para melhorar o aprendizado das redes neurais sobre este problema.

show abstract