摘要
提出了一种基于Rough集和RBF神经网络结合的车牌字符识别方法。该方法针对车牌字符二值化图像,给出了基于粗糙集理论的知识获取方法,包括根据训练样本的特征向量建立决策表、离散决策表属性、约简决策表属性,然后由约简后的属性构造RBF网络识别器。试验表明该方法有效地减少了决策属性的个数,简化了神经网络识别器的结构,提高了泛化能力和抗噪声能力,在车牌字符识别中取得了较好的识别效果。
A method based on the combination of rough set and Radial Basis Function （RBF） neural Network is presented for the car＇s plate characters recognition. Directing to the car＇s plate character recognition, this paper introduces the knowledge acquisition method based on rough set theory. It includes construct decision table, attribute discretization and attribute reduction. Finally, the reduced decision attributes are used to construct neural network recognizing machine. The method can reduce the numbers of attributes in the decision table, simplify the structure of neural network and improve the ability of generality. The experiment results of the car＇s plate character recognition show that the algorithms are practical and effective.
出处
《安徽建筑工业学院学报（自然科学版）》
2006年第4期87-90,共4页
Journal of Anhui Institute of Architecture(Natural Science)
基金
安徽省教育厅自然科学项目(2006KJ013C)
