摘要
条码的分类检测对条码识别具有非常重要的意义.本文提出将基于支持向量机的多分类方法用于条码的分类检测.对每种条码采用支持向量机二值分类器进行分类,这些二值分类器组成决策树的节点,构成决策树.通过实验表明,SVM在样本有限的情况下具有非常好的泛化能力.
Barcode classification is very important in Barcode recognition.In this paper,the method of multi-class classification based on Support Vector Machines was used in barcode classification.Each group of barcode was classified by a SVM classifier.These binary classifiers were seen as the node of decision tree to construct a decision classifying tree.From the experiment,it is proved that the SVM has better generalization ability under the conditions of limited samples.
出处
《山东师范大学学报(自然科学版)》
CAS
2007年第4期24-26,共3页
Journal of Shandong Normal University(Natural Science)
基金
山东省自然科学基金资助项目(Z2004G02)