【摘要】 在生产香烟包装纸时,需要将不合格产品剔除,同时需要知道包装产品的优良率。针对香烟包装图案的特点,介绍了一种利用RBF网络在线检测和诊断香烟包装质量状况的方法。首先,采集具有典型特征的样本,建立足够数量的样本特征集。然后设计一个合适的RBF网络结构,采用样本集对网络进行训练和测试,最终得到一个实用的RBF诊断网络。实验表明,该系统完全能够满足质量检测的要求,可用于包装质量的在线诊断。更多还原

【Abstract】 In the process of producing cigarette package,we need to know the fine rate of the package product and eliminate the unqualified product.In view of the feature of cigarette package design,this paper introduces a kind of online detection and diagnosis method of package design quality based on the RBF neural network.First,we must gather the typical samples and establish enough sample feature sets.And then we design a appropriate RBF network,train and test it by using the sample feature sets until it is available.The experiment has showed that this system can definitely satisfy the request of the quality detection,and may be used in online detection and diagnosis of cigarette package quality.更多还原