摘要
目前人工神经网络广泛的应用于各种模式识别以及自动控制等系统中,该系统利用了图像去噪,分割,倾斜度调整,字符分割,字符归一化等图像预处理之后,将人民币号码图像进行特征提取,应用神经网络进行识别.同时,用遗传算法对人工神经网络进行网络权值的优化,加快了训练速度,提高了识别率.
Now, Neural Networks have been applied to many kinds of pattern recognition and autocontrol system widely. RMB number recognition system presented in this paper used noise elimination, image segmentation, gradient emendation, characters partition, and size standardization to pretreat the image. Then, the system extracted the features. Neural Networks are used to recognize the letters and numbers of RMB. In this system, Genetic algorithm is used to optimize the networks, by this way,we can shorten the training time effectively and increase the recognition rate.
出处
《四川大学学报（自然科学版）》
CAS
CSCD
北大核心
2007年第4期812-816,共5页
Journal of Sichuan University(Natural Science Edition)
关键词
人工神经网络
遗传算法
特征提取
neural networks, genetic algorithm, feature extraction
