Análise do desempenho de modelos de optical character recognition de propósito geral em placas veiculares
Resumo
O Reconhecimento Automático de Placas Veiculares (ALPR) é uma tecnologia importante para diversas áreas, como sistemas de transporte, cujo desenvolvimento pode apresentar altos custos e complexidade. Nesse contexto, modelos de Optical Character Recognition (OCR) de propósito geral emergem como uma potencial alternativa. Este trabalho realizou uma avaliação quantitativa de três OCRs genéricos (EasyOCR, Pytesseract, Keras-OCR). Os resultados revelam uma discrepância de desempenho: modelos estado da arte atingem 95,6% de acurácia, enquanto o OCR genérico de melhor performance alcançou somente 16,5%. Tais achados demonstram que ferramentas de OCR não especializadas carecem de adequação para aplicações reais em ALPR.Referências
Atienza, R. (2021). Vision transformer for fast and efficient scene text recognition. In International conference on document analysis and recognition, pages 319–334. Springer.
Baek, J., Kim, G., Lee, J., Park, S., Han, D., Yun, S., Oh, S. J., and Lee, H. (2019). What is wrong with scene text recognition model comparisons? dataset and model analysis. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4715–4723.
Bulan, O., Kozitsky, V., and Burry, A. (2015). Towards annotation free license plate recognition. In 2015 IEEE 18th International Conference on Intelligent Transportation Systems, pages 1495–1499. IEEE.
Du, S., Ibrahim, M., Shehata, M., and Badawy, W. (2012). Automatic license plate recognition (alpr): A state-of-the-art review. IEEE Transactions on circuits and systems for video technology, 23(2):311–325.
Jaided AI (2024). Easyocr: Ready-to-use ocr with 80+ supported languages. [link]. Acessado em 24 de agosto de 2025.
Laroca, R., Cardoso, E. V., Lucio, D. R., Estevam, V., and Menotti, D. (2022). On the cross-dataset generalization in license plate recognition. In International Conference on Computer Vision Theory and Applications (VISAPP), pages 166–178.
Laroca, R., Zanlorensi, L. A., Estevam, V., Minetto, R., and Menotti, D. (2023). Leveraging model fusion for improved license plate recognition. In Iberoamerican Congress on Pattern Recognition (CIARP), pages 60–75.
Montazzolli, S. and Jung, C. (2017). Real-time brazilian license plate detection and recognition using deep convolutional neural networks. In 2017 30th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 55–62.
Morales, F. (2020). Keras-ocr: End-to-end trainable ocr pipeline. [link]. Acessado em 24 de agosto de 2025.
Shi, B., Bai, X., and Yao, C. (2016). An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition. IEEE transactions on pattern analysis and machine intelligence, 39(11):2298–2304.
Tesseract-OCR. (2024). Tesseract open source ocr engine. [link]. Acessado em 24 de agosto de 2025.
Baek, J., Kim, G., Lee, J., Park, S., Han, D., Yun, S., Oh, S. J., and Lee, H. (2019). What is wrong with scene text recognition model comparisons? dataset and model analysis. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4715–4723.
Bulan, O., Kozitsky, V., and Burry, A. (2015). Towards annotation free license plate recognition. In 2015 IEEE 18th International Conference on Intelligent Transportation Systems, pages 1495–1499. IEEE.
Du, S., Ibrahim, M., Shehata, M., and Badawy, W. (2012). Automatic license plate recognition (alpr): A state-of-the-art review. IEEE Transactions on circuits and systems for video technology, 23(2):311–325.
Jaided AI (2024). Easyocr: Ready-to-use ocr with 80+ supported languages. [link]. Acessado em 24 de agosto de 2025.
Laroca, R., Cardoso, E. V., Lucio, D. R., Estevam, V., and Menotti, D. (2022). On the cross-dataset generalization in license plate recognition. In International Conference on Computer Vision Theory and Applications (VISAPP), pages 166–178.
Laroca, R., Zanlorensi, L. A., Estevam, V., Minetto, R., and Menotti, D. (2023). Leveraging model fusion for improved license plate recognition. In Iberoamerican Congress on Pattern Recognition (CIARP), pages 60–75.
Montazzolli, S. and Jung, C. (2017). Real-time brazilian license plate detection and recognition using deep convolutional neural networks. In 2017 30th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 55–62.
Morales, F. (2020). Keras-ocr: End-to-end trainable ocr pipeline. [link]. Acessado em 24 de agosto de 2025.
Shi, B., Bai, X., and Yao, C. (2016). An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition. IEEE transactions on pattern analysis and machine intelligence, 39(11):2298–2304.
Tesseract-OCR. (2024). Tesseract open source ocr engine. [link]. Acessado em 24 de agosto de 2025.
Publicado
12/11/2025
Como Citar
TOZEVICH, Luís Gustavo Werle; LUNARDI, Gabriel Machado; SILVEIRA, Thiago Lopes Trugillo da; OLIVEIRA, Adriano Quilião de.
Análise do desempenho de modelos de optical character recognition de propósito geral em placas veiculares. In: ESCOLA REGIONAL DE APRENDIZADO DE MÁQUINA E INTELIGÊNCIA ARTIFICIAL DA REGIÃO SUL (ERAMIA-RS), 1. , 2025, Porto Alegre/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 352-355.
DOI: https://doi.org/10.5753/eramiars.2025.16717.
