基于支持向量机的字符编码识别系统

基于支持向量机的字符编码识别系统 The Design of Recognition System for Numbers of Detonators Based on Support Vector Machine 孙孟方 SUN Mengfang1 1 first-author 张立文 ZHANG Liwen2 2 高火涛 GAO Huotao3 3 中国航空工业集团公司洛阳电光设备研究所 中国航空工业集团公司洛阳电光设备研究所 Luoyang Institute of Electro-Optical Equipment,Aivation Industry Corporation of China Luoyang Institute of Electro-Optical Equipment,Aivation Industry Corporation of China 河南科技大学电子信息工程学院 河南科技大学电子信息工程学院 School of Electronic Information Engineering,Henan University of Science and Technology School of Electronic Information Engineering,Henan University of Science and Technology 武汉大学电子信息学院 武汉大学电子信息学院 School of Electronic Information,Wuhan University School of Electronic Information,Wuhan University 基于支持向量机设计了一种产品字符编码识别系统,该系统通过CCD视觉传感器采集图像信息,经过目标提取、字符分割、编码识别过程,最后输出识别结果.其中,识别过程采用支持向量机作为判别函数分类器,该方法能较好的解决小样本、非线性、高维数等实际问题,并且较传统的神经网络识别方法训练速度更快.实验结果表明,该方法识别率高,可以达到98.3%,并且具有较高的实时性. We design an identification system for numbers.First,the system gets image information by vision sensor CCD.The results can be achieved by the process of target extraction,character segmentation,code identification;the recognition process use support vector machine（SVM） as a discriminant function classifier.The method can better solve the actual problems such as small samples,nonlinear,high dimension and so on.Moreover,the proposed method in this paper is much faster than the traditional methods such as the neural network method.Experimental results show that the method developed in this paper can achieve a higher recognition rate over 98.3% and high real-time. 支持向量机 编码识别 图像处理 support vector machine（SVM） number recognition image processing 10.14188/j.1671-8836.2013.03.005 TP391.41 河南省重点科技攻关项目（112102210430）资助 2013-03-01 2021-04-01 3 59