【作者】 苑玮琦； 穆长江； 李德胜；

【Author】 Yuan Weiqi Mu Changjiang Li Desheng(Computer vision Group, Shenyang University of Technology, Shenyang 110023, China) (Dep. of Automatic Measurement and Control, Harbin Institute of Technology, Harbin 150001, China)

【机构】 沈阳工业大学视觉检测技术研究所； 哈尔滨工业大学自动化测试与控制系；

【摘要】 提出了一种基于汉字结构特征的车牌照字符分割方法。该方法首先找到汉字的主体部件,然后根据汉字上下、左右、包围结构找到组成汉字的其它部件,并结合汉字识别结果对分割得到的汉字进行确定。最后,根据车牌字符串的结构特征找到其它字符。特殊车牌图像,要对字符和字符、字符和边界粘连的情况进行分裂,对部分字符进行合并。而且本方法还可以对定位出现车牌字符丢失情况进行重新定位,找到丢掉的字符。对大量车牌图像字符的分割,试验结果证明,该方法能够达到准确分割出车牌图像中字符的目的。更多还原

【Abstract】 This page puts forward a vehicle plate character segmentation method based on Chinese character. First,this method finds the main parts of the Chinese character,then finds another compositive parts of the Chinese character based on up-down,left-right,enveloping structures of the Chinese character,and then combines with character recognition result to determine the segmented Chinese character. Finally, this method finds another characters based on the strucure features of vehicle plate string. As for special vehicle plate image,this method must split these characters when they conglutinate together with other characters or the edges of vehicle plate, and merge the parts of character. In addition,this method locates the characters over again and finds the losed characters when vehicle plate characters lose while locating. We have segmented a great deal characters of vehicle plate image,the experiment results show that the proposed method can gain the goal of segmentation the characters of vehicle image well and truly.更多还原