Object Detection in Container Terminals Based on Deep Learning Approach: A Systematic Literature Review

After the pandemic, the container trade experienced a significant increase. This increase has resulted in several ports and container terminals facing operational problems. To deal with these problems, some operations at container terminals have been carried out automatically. One requirement for automated operations at container ports is the ability to identify objects inside the terminal's environment automatically. One aspect of computer vision, image detection, has been widely applied in security and health. With image detection, the process of identifying and detecting an object can be done in real time and accurately. This paper aims to review previous studies discussing the topic of object detection in container terminals. The main focus of prior research on object detection based on one of the widely used approaches, namely deep learning, is systematically presented in this study. According to previous research, the most frequently detected objects were containers or parts of containers.