摘要
针对已有集装箱箱号字符定位、分割、识别算法上的不足,分别提出了基于逆向扫描的箱号区域提取法,基于区域重心的字符定位分割法和基于概率神经网络的字符识别法。用MATLAB编程,实现了对集装箱箱号的自动识别,识别正确率可达95%。提出的方法具有识别正确率高、算法简单、可靠性好等优点,有望提高码头道口通勤率,降低运营成本。
In view of the insufficiency of the existing positioning, segmentation and recognition algorithm to container codes, the codes area extraction method based on reverse scan, character location and segmentation method based on the center of gravity of area, character recognition method based on probabilistic neural network are presented. It implements automatic identification of the container codes using MATLAB, the recognition accuracy up to 95%. The proposed method has the advantages of high recognition accuracy, simple algorithm and good reliability. So it is expected to increase commuting rate of terminal crossing, and reduce operating costs.
出处
《电视技术》
北大核心
2015年第19期109-112,共4页
Video Engineering
基金
福建省科技厅资助省属高校专项(JK2014024)
关键词
集装箱箱号
逆向扫描
字符分割
概率神经网络
字符识别
container codes
reverse scan
character segmentation
probabilistic neural network
character recognition
