Study on the impact of the degradation method on the generalization of Super-Resolution models for ALPR
Resumo
Dada a complexidade das variações nos cenários e nos equipamentos, é fundamental empregar métodos avançados de aprimoramento de imagens no Reconhecimento Automático de Placas de Licença (ALPR). Este estudo analisou o impacto de diferentes métodos de degradação de imagens na síntese de dados para treinamento de modelos baseados na arquitetura de super-resolução Real-ESRGAN. Os resultados demonstraram um poder de generalização significativamente maior ao utilizar um conjunto de dados construído com um método de degradação mais robusto.
Palavras-chave:
Reconhecimento Automático de Placas de Veículos, Aprimoramento de Imagem, Super-Resolução, Métodos de Degradação de Imagem, Generalização de Modelos
Referências
Abdelaziz, A. H., Chan, Y. K., and Koo, V. C. (2021). Enhancement for license plate recognition using image super resolution technique. In 2021 International Conference on Electrical, Communication, and Computer Engineering (ICECCE), pages 1–4. IEEE.
Asaad, A., Faizabadi, A. R., and Zaki, H. F. M. (2023). Synthetic license plate generation: A novel approach for effective license plate recognition in malaysia. In 2023 IEEE 8th International Conference on Engineering Technologies and Applied Sciences (ICETAS), pages 1–8. IEEE.
Chen, H., He, X., Qing, L., Wu, Y., Ren, C., Sheriff, R. E., and Zhu, C. (2022). Real-world single image super-resolution: A brief review. Information Fusion, 79:124–145.
Hamdi, A., Chan, Y. K., and Koo, V. C. (2021). A new image enhancement and super resolution technique for license plate recognition. Heliyon, 7(11).
Kim, D., Kim, J., and Park, E. (2024). Afa-net: Adaptive feature attention network in image deblurring and super-resolution for improving license plate recognition. Computer Vision and Image Understanding, 238:103879.
Laroca, R., Severo, E., Zanlorensi, L. A., Oliveira, L. S., Gonçalves, G. R., Schwartz, W. R., and Menotti, D. (2018). A robust real-time automatic license plate recognition based on the yolo detector. In 2018 international joint conference on neural networks (ijcnn), pages 1–10. IEEE.
Nascimento, V., Laroca, R., Lambert, J. d. A., Schwartz, W. R., and Menotti, D. (2022). Combining attention module and pixel shuffle for license plate super-resolution. In 2022 35th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), volume 1, pages 228–233. IEEE.
Nascimento, V., Laroca, R., Lambert, J. d. A., Schwartz, W. R., and Menotti, D. (2023). Super-resolution of license plate images using attention modules and sub-pixel convolution layers. Computers & Graphics, 113:69–76.
Pan, S., Chen, S.-B., and Luo, B. (2023). A super-resolution-based license plate recognition method for remote surveillance. Journal of Visual Communication and Image Representation, 94:103844.
Pan, Y., Tang, J., and Tjahjadi, T. (2024). Lpsrgan: Generative adversarial networks for super-resolution of license plate image. Neurocomputing, page 127426.
Pattanaik, A. and Balabantaray, R. C. (2023). Enhancement of license plate recognition performance using xception with mish activation function. Multimedia tools and applications, 82(11):16793–16815.
Pourhadi, N., Shafizadeh, B., Eshghi, F., and Kelarestaghi, M. (2022). Yolov5-based alpr improvement using selective sr-gan. In 2022 2nd International Conference on Computing and Machine Intelligence (ICMI), pages 1–6. IEEE.
Rao, Z., Yang, D., Chen, N., and Liu, J. (2024). License plate recognition system in unconstrained scenes via a new image correction scheme and improved crnn. Expert Systems with Applications, 243:122878.
Sereethavekul, W. and Ekpanyapong, M. (2023). Adaptive lightweight license plate image recovery using deep learning based on generative adversarial network. IEEE Access, 11:26667–26685.
Wang, H., Li, Y., Dang, L.-M., and Moon, H. (2021a). Robust korean license plate recognition based on deep neural networks. Sensors, 21(12):4140.
Wang, X., Xie, L., Dong, C., and Shan, Y. (2021b). Real-esrgan: Training real-world blind super-resolution with pure synthetic data. In Proceedings of the IEEE/CVF international conference on computer vision, pages 1905–1914.
Yang, S., Xie, L., Ran, X., Lei, J., and Qian, X. (2024). Pragmatic degradation learning for scene text image super-resolution with data-training strategy. Knowledge-Based Systems, 285:111349.
Asaad, A., Faizabadi, A. R., and Zaki, H. F. M. (2023). Synthetic license plate generation: A novel approach for effective license plate recognition in malaysia. In 2023 IEEE 8th International Conference on Engineering Technologies and Applied Sciences (ICETAS), pages 1–8. IEEE.
Chen, H., He, X., Qing, L., Wu, Y., Ren, C., Sheriff, R. E., and Zhu, C. (2022). Real-world single image super-resolution: A brief review. Information Fusion, 79:124–145.
Hamdi, A., Chan, Y. K., and Koo, V. C. (2021). A new image enhancement and super resolution technique for license plate recognition. Heliyon, 7(11).
Kim, D., Kim, J., and Park, E. (2024). Afa-net: Adaptive feature attention network in image deblurring and super-resolution for improving license plate recognition. Computer Vision and Image Understanding, 238:103879.
Laroca, R., Severo, E., Zanlorensi, L. A., Oliveira, L. S., Gonçalves, G. R., Schwartz, W. R., and Menotti, D. (2018). A robust real-time automatic license plate recognition based on the yolo detector. In 2018 international joint conference on neural networks (ijcnn), pages 1–10. IEEE.
Nascimento, V., Laroca, R., Lambert, J. d. A., Schwartz, W. R., and Menotti, D. (2022). Combining attention module and pixel shuffle for license plate super-resolution. In 2022 35th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), volume 1, pages 228–233. IEEE.
Nascimento, V., Laroca, R., Lambert, J. d. A., Schwartz, W. R., and Menotti, D. (2023). Super-resolution of license plate images using attention modules and sub-pixel convolution layers. Computers & Graphics, 113:69–76.
Pan, S., Chen, S.-B., and Luo, B. (2023). A super-resolution-based license plate recognition method for remote surveillance. Journal of Visual Communication and Image Representation, 94:103844.
Pan, Y., Tang, J., and Tjahjadi, T. (2024). Lpsrgan: Generative adversarial networks for super-resolution of license plate image. Neurocomputing, page 127426.
Pattanaik, A. and Balabantaray, R. C. (2023). Enhancement of license plate recognition performance using xception with mish activation function. Multimedia tools and applications, 82(11):16793–16815.
Pourhadi, N., Shafizadeh, B., Eshghi, F., and Kelarestaghi, M. (2022). Yolov5-based alpr improvement using selective sr-gan. In 2022 2nd International Conference on Computing and Machine Intelligence (ICMI), pages 1–6. IEEE.
Rao, Z., Yang, D., Chen, N., and Liu, J. (2024). License plate recognition system in unconstrained scenes via a new image correction scheme and improved crnn. Expert Systems with Applications, 243:122878.
Sereethavekul, W. and Ekpanyapong, M. (2023). Adaptive lightweight license plate image recovery using deep learning based on generative adversarial network. IEEE Access, 11:26667–26685.
Wang, H., Li, Y., Dang, L.-M., and Moon, H. (2021a). Robust korean license plate recognition based on deep neural networks. Sensors, 21(12):4140.
Wang, X., Xie, L., Dong, C., and Shan, Y. (2021b). Real-esrgan: Training real-world blind super-resolution with pure synthetic data. In Proceedings of the IEEE/CVF international conference on computer vision, pages 1905–1914.
Yang, S., Xie, L., Ran, X., Lei, J., and Qian, X. (2024). Pragmatic degradation learning for scene text image super-resolution with data-training strategy. Knowledge-Based Systems, 285:111349.
Publicado
17/11/2024
Como Citar
OLIVEIRA, Cristiano L.; MATOS, Leonardo N.; M. NETO, Paulo S. G. de.
Study on the impact of the degradation method on the generalization of Super-Resolution models for ALPR. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 21. , 2024, Belém/PA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2024
.
p. 613-624.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2024.245088.