A Novel Framework for Container Code-Character Recognition Based on Deep Learning and Template Matching

Effective and efficient automatic container code recognition (ACCR) is an important part of Intelligent port management. Due to the severity of nonuniform illumination and invalidation of color information, higher sensitivity and intelligence are required in ACCR. To address this issue, a novel Framework for container code-character recognition based on deep learning and template matching (TM) in visual images is proposed in this paper. In this framework, convolutional neural network (CNN) algorithm and TM algorithm are applied to recognize the container code-character, respectively. Then, the container code is obtained by combining all characters recognized according to the credibility of each character and the verification rule of container code. The experimental results demonstrate the effectiveness of the proposed method, which can successfully recognize the container code in complicated situations.