FoodLens AI: Um Sistema de Reconhecimento de Produtos com Visão Computacional, OCR e Enriquecimento de Dados em Tempo Real
Resumo
Este artigo apresenta o FoodLens AI, um sistema mobile para reconhecimento de produtos em tempo real utilizando Visão Computacional e OCR. O sistema identifica produtos embalados e extrai informações textuais dos rótulos. A arquitetura integra detecção de objetos com YOLOv8 (avaliada empiricamente em cenários reais de varejo), e prevê etapas de pré-processamento de imagem e OCR para extração textual dos rótulos. Os dados extraídos são enriquecidos com fontes externas e LLMs. Diferente de abordagens focadas apenas em alimentos, o sistema amplia a compreensão de produtos. Além da análise nutricional, oferece interpretação de ingredientes, comparação entre produtos e informações de mercado, auxiliando a tomada de decisão do consumidor.
Referências
(2024). Ocr and llm-based analysis of packaged food labels. Frontiers in Health Informatics, 13(3).
(2025). Optimization of tesseract ocr for automatic text extraction. IJSECS.
Amugongo, L. M. et al. (2023). Mobile computer vision-based applications for food recognition. IEEE Access.
Arkin, E., Yadikar, N., Xu, X., Aysa, A., and Ubul, K. (2023). A survey: object detection methods from CNN to transformer. Multimedia Tools and Applications, 82(14):21353–21383.
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., and Zagoruyko, S. (2020). End-to-end object detection with transformers. In Computer Vision – ECCV 2020, volume 12346 of Lecture Notes in Computer Science, pages 213–229. Springer.
Domiciano, M. (2025). Image pre-processing techniques for ocr. [link].
Guru, S. et al. (2024). Implementation of health impact assessment of packaged foods through nutritional label recognition using ocr. Journal of Food Informatics.
Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics YOLO. [link].
Krizhevsky, A., Sutskever, I., and Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, volume 25.
Label Your Data (2026). Ocr deep learning: How do you do it in 2026. [link].
Le, T.-T.-H. et al. (2025). A robust framework for coffee bean package label recognition. Sensors.
Liu, C., Cao, Y., Luo, Y., Chen, G., Vokkarane, V., and Ma, Y. (2016). DeepFood: Deep learning-based food image recognition for computer-aided dietary assessment. In Inclusive Smart Cities and Digital Health, volume 9677 of Lecture Notes in Computer Science, pages 37–48. Springer.
Lv, W., Zhao, Y., Xu, S., Wei, J., Wang, G., Cui, C., Du, Y., Dang, Q., and Liu, Y. (2023). DETRs beat YOLOs on real-time object detection.
Mezgec, S. and Koroušić Seljak, B. (2017). NutriNet: A deep learning food and drink image recognition system for dietary assessment. Nutrients, 9(7):657.
Mohbat, F., Zaki, M. J., et al. (2025). Extract nutritional information from bilingual food labels using large language models. Journal of Imaging, 11(8).
Open Food Facts (2024). Collective achievements 2024. [link].
Saputra, M. E., Susanto, A., and Carmelita, B. J. (2024). Implementation of tesseract ocr and bounding box for text extraction on food nutrition labels. International Journal of Advanced Computer Science and Applications.
Seitaj, H. and Elangovan, V. (2024). Information extraction from product labels. International Journal of Artificial Intelligence and Applications (IJAIA), 15(2).
Wang, C.-Y., Liao, H.-Y. M., Wu, Y.-H., Chen, P.-Y., Hsieh, J.-W., and Yeh, I.-H. (2020). CSPNet: A new backbone that can enhance learning capability of CNN. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 390–391.
(2025). Optimization of tesseract ocr for automatic text extraction. IJSECS.
Amugongo, L. M. et al. (2023). Mobile computer vision-based applications for food recognition. IEEE Access.
Arkin, E., Yadikar, N., Xu, X., Aysa, A., and Ubul, K. (2023). A survey: object detection methods from CNN to transformer. Multimedia Tools and Applications, 82(14):21353–21383.
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., and Zagoruyko, S. (2020). End-to-end object detection with transformers. In Computer Vision – ECCV 2020, volume 12346 of Lecture Notes in Computer Science, pages 213–229. Springer.
Domiciano, M. (2025). Image pre-processing techniques for ocr. [link].
Guru, S. et al. (2024). Implementation of health impact assessment of packaged foods through nutritional label recognition using ocr. Journal of Food Informatics.
Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics YOLO. [link].
Krizhevsky, A., Sutskever, I., and Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, volume 25.
Label Your Data (2026). Ocr deep learning: How do you do it in 2026. [link].
Le, T.-T.-H. et al. (2025). A robust framework for coffee bean package label recognition. Sensors.
Liu, C., Cao, Y., Luo, Y., Chen, G., Vokkarane, V., and Ma, Y. (2016). DeepFood: Deep learning-based food image recognition for computer-aided dietary assessment. In Inclusive Smart Cities and Digital Health, volume 9677 of Lecture Notes in Computer Science, pages 37–48. Springer.
Lv, W., Zhao, Y., Xu, S., Wei, J., Wang, G., Cui, C., Du, Y., Dang, Q., and Liu, Y. (2023). DETRs beat YOLOs on real-time object detection.
Mezgec, S. and Koroušić Seljak, B. (2017). NutriNet: A deep learning food and drink image recognition system for dietary assessment. Nutrients, 9(7):657.
Mohbat, F., Zaki, M. J., et al. (2025). Extract nutritional information from bilingual food labels using large language models. Journal of Imaging, 11(8).
Open Food Facts (2024). Collective achievements 2024. [link].
Saputra, M. E., Susanto, A., and Carmelita, B. J. (2024). Implementation of tesseract ocr and bounding box for text extraction on food nutrition labels. International Journal of Advanced Computer Science and Applications.
Seitaj, H. and Elangovan, V. (2024). Information extraction from product labels. International Journal of Artificial Intelligence and Applications (IJAIA), 15(2).
Wang, C.-Y., Liao, H.-Y. M., Wu, Y.-H., Chen, P.-Y., Hsieh, J.-W., and Yeh, I.-H. (2020). CSPNet: A new backbone that can enhance learning capability of CNN. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 390–391.
Publicado
23/09/2026
Como Citar
SANTOS, Stheffany; DURAN, Pedro; SANTOS, Davi Rocha dos; ABREU, Pedro; ARAUJO, Victor Flávio.
FoodLens AI: Um Sistema de Reconhecimento de Produtos com Visão Computacional, OCR e Enriquecimento de Dados em Tempo Real. In: ESCOLA REGIONAL DE COMPUTAÇÃO BAHIA, ALAGOAS E SERGIPE (ERBASE), 26. , 2026, Maragogi/AL.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 180-189.
DOI: https://doi.org/10.5753/erbase.2026.29908.
