摘要
为实现对集装箱箱号的正确识别,提出一种基于模板匹配和特征匹配的识别算法。对采集到的集装箱图像进行预处理,得到改善后的集装箱二值化图像;采用数学形态学操作使字符域连通,计算字符连通域的宽高比得到集装箱箱号区域;利用投影检测方法实现对箱号字符的分割;运用模板匹配算法与特征匹配算法相结合的分类方法对集装箱箱号字符进行识别。该算法用MATLAB进行编程,完成对集装箱箱号的自动定位、分割和识别。提出的方法可正确识别出集装箱箱号,识别率达到93%,识别时间为130~150 ms,可提高码头的工作效率。
In order to achieve correct recognition of container code,a recognition algorithm based on template matching and feature matching is proposed. The collected container pictures are preprocessed to obtain the improved container binary images; the mathematical morphological operation is used to connect the character domain,and the aspect ratio of the connected domain is calculated to get the container code area; the projection detection method is used to achieve character segmentation; the classification method combining the template matching algorithm and the feature matching algorithm is used to recognize the container code characters. The algorithm is programmed with MATLAB to achieve automatic positioning,segmentation and recognition of container codes. The proposed method correctly recognizes container codes,the recognition rate reaches 93%,and the identification time is between 130 ms and 150 ms. The method can improve the working efficiency of terminals.
引文
[1]宓超,沈阳,宓为建.装卸机器视觉及应用[M].上海:上海科学技术出版社,2016:147-151.
[2]胡婷.基于神经网络的集装箱字符识别的研究[D].武汉:武汉理工大学,2012.
[3]安博文,李丹,庞然.基于SVM分类器的集装箱箱号识别法[J].上海海事大学学报,2011,32(1):25-29.
[4]孙凌红.集装箱箱号智能识别算法研究[D].武汉:武汉理工大学,2012.
[5]陈子宜.基于机器视觉的集装箱箱号识别[D].上海:上海交通大学,2014.
[6]张铮,徐超,任淑霞,等.数字图像处理与机器视觉[M].2版.北京:人民邮电出版社,2014:58-89.
[7]孙正.数字图像处理与识别[M].北京:机械工业出版社,2014:69-74.
[8]陈莉.数字图像处理算法研究[M].北京:科学出版社,2016:97-113.
[9] GONZALEZ R C,WOODS R E.数字图像处理[M].阮秋琦,阮宇智,译.3版.北京:电子工业出版社,2011:402-437.
[10]谷秋頔.基于模板匹配的车牌字符识别及其判别函数的研究[D].太原:中北大学,2012.
[11]冼允廷,路小波,施毅,等.基于投影二分法的车牌字符分割方法[J].交通与计算机,2007(5):69-72.
[12]陈默,何小海,吴炜,等.结合独立与连续字符识别的集装箱号识别技术[J].四川大学学报(工程科学版),2011,43(S1):39-145.DOI:10.15961/j.jsuese.2011.s1.025.
[13]姜瑾,张桂林,许慧慧.基于投影特征曲线匹配的车牌字符识别算法[J].计算机与数字工程,2007(6):20-22.
[14]陈丹.集装箱图像识别与定位系统研究与实现[D].成都:西南交通大学,2013.