大噪声图像的轮胎规格号识别技术  

A Recognition Technology of Tire Specifications of Large Noise Image

作　　者：杨基春[1] 黄战华[1] 蔡怀宇[1] 张尹馨[1] 

机构地区：[1]天津大学精仪学院光电信息工程系,天津300072

出　　处：《光电工程》2010年第9期98-103,共6页Opto-Electronic Engineering

摘　　要：针对轮胎规格号图像高噪声的特点,本文提出字符截断方式来获得可区分的外轮廓特征,利用BP神经网络并结合规格号关系特征来识别轮胎规格号。首先,获取可区分规格号字符样本特征:截取字符的三分之二,获取其外轮廓游程特征;随后,对其进行BP神经网络训练和识别,得到一次识别结果;最后,采用规格号类型特征,对识别结果做二次识别。实验结果表明,BP神经网络对变形字符有较强的容错能力,特征获取和二次识别算法能够有效识别轮胎规格号。For big noises tire image, a method of character truncation was presented to get distinguishing outer contour features. Artificial neural network and features of relations was used to recognize tire specifications. First, the sample features of distinguishing specifications were gotten： two-thirds characters were intercepted to get run-length characteristics of its external contour. Then, one recognition result was gotten by BP neural network training and recognition. Finally, character features of relations processed fore results were applied to do the second recognition. Experimental results show that the characteristics of acquisition and the second recognition algorithms can effectively identify the tire specifications. Moreover, BP neural network has better fault-tolerance and recognition rate for deformation characters caused by noises, and feature gotten and secondary identification algorithm can effectively identify tire specifications.

关 键 词：BP神经网络 轮胎规格号 字符截断 关系特征 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]