Maritime transport plays a pivotal role in global trade, where efficiency and 1 accuracy in port operations are crucial. An aspect of the large scale port operations is 2 container code recognition. Manual inspection of container codes leads to delays and 3 errors, prompting the need for intelligent, automated solutions. This article proposes a 4 hybrid approach to Optical Character Recognition systems that integrates YOLOv7 for text 5 detection with the TrOCR transformer-based model for the recognition of the container 6 codes alphanumerical characters. The systems are designed to address key challenges 7 encountered in real-world port environments, such as varying lighting conditions, image 8 distortions and variable container code orientations. A comprehensive evaluation was 9 conducted utilizing datasets that reflect the diverse conditions of port operations, including 10 different orientations and lighting scenarios. The results indicate that this hybrid approach 11 achieves high precision in both text detection and recognition tasks, outperforming tra- 12 ditional OCR tools like Tesseract and EasyOCR. The integration of YOLOv7 and TrOCR 13 allows for end-to-end processing with improved robustness and accuracy. By handling 14 multi-orientation container codes, the proposed system demonstrates an advancement in 15 automatic code recognition, offering a scalable solution for enhancing operational efficiency 16 in port and freight logistics.
Santos et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: