Automation in Detecting Deviations in Carton Stamping to Ensure Product Integrity and Safety

• Claudio Honorio da Silva Junior Faculdade Biopark
• Mateus Heck Soares Faculdade Biopark
• Leonardo Garcia Tampelini Faculdade Biopark
• Jhoni Eldor Schulz Faculdade Biopark

DOI: https://doi.org/10.5753/latinoware.2024.245733

Abstract

This paper investigates an approach to enhance quality control in the pharmaceutical industry by leveraging optical character recognition (OCR) techniques for automated batch identification on drug packaging. The current manual inspection process is susceptible to human errors and delays, jeopardizing the accuracy and safety of drug production. To address these concerns, this study explores the application of artificial neural networks (RNAs) to automate the verification process.

Keywords: Optical Character Recognition (OCR), Machine Learning, Artificial Neural Networks (RNAs), Image Preprocessing

References

Jan Erik Solem (2012). Programming Computer Vision with Python.

Jaided AI. Easy OCR documentation [link]

Pytorch. Pytorch documentation [link]

Tschida, C. (2019). Designing for Accessibility: A Complete Guide to Creating Inclusive User Experiences.

Videojet. (2023). Videojet 1880HR. Videojet Brasil. [link]

Li, M., Lv, T., Chen, J., Cui, L., Lu, Y., Florencio, D., Zhang C., Li Z., Wei, F. (2022). TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models. arXiv preprint arXiv:2109.10282.

TURNER, T. Pattern Recognition in Image Analysis. Cambridge: Cambridge University Press, 2017.

MARTIN, M. Image Segmentation Techniques. Oxford: Oxford University Press,2016.