Container code recognition using computer vision and support vector machines

Abstract

An automated container-code recognition plays an important role in modern container management system. Many similar techniques have been applied for the automatic license plate recognition (ALPR) system. Compared to the ALPR system, container-code recognition faces more problems due to the severity of nonuniform illumination and invalidation of color information. The system in this paper consists of two main parts, namely region detection and character recognition. The region detection module uses scanlines to search for the most pixels to locate the character region. The recognition module use HOG features extracted from the image for trained support vector machines (SVMs) system. The experimental results demonstrate the efficiency and effectiveness of the proposed technique for practical usage.