A Vision-based Container-Code Checking System: Case Study at International Terminal
Duc-Dat Ngo, Van-Hoang-Anh Phan, Huynh-The Pham, Tien-Tan Be, Van-Binh Nguyen, My-Ha Le · 2023
Checking the container code at an international terminal is necessary to ensure safety, security, and compliance in the global trade ecosystem. By implementing such a system, ports can mitigate risks, prevent illegal activities, and facilitate the smooth flow of legitimate cargo while protecting the interests of all stakeholders involved. This paper introduces a vision-based container-code-checking system built on a versatile and reasonably priced hardware platform with promising performance. More concretely, the proposed system includes two different stages executed continuously. To begin with, the CRAFT is packaged in EasyOCR to detect the container-code area. Subsequently, a Convolutional Neural Network combines with Spatial Transformer Network to classify each character of the detected container-code area. From that, our system facilitates data collecting and analysis related to container movements, inspections, and compliance. The demo video can be watched here: https://www.youtube.com/watch?v=JsRZEUk_lfMl