摘要
针对输送线烟箱缺条检测时存在的检测精度较低、抗噪性较差问题,探讨关联规则挖掘方法在输送线烟箱缺条智能检测中的应用。以关联规则挖掘概念为依据,结合特征匹配度概念,获取输送线烟箱全部特征后构建烟箱缺条特征库;采用烟箱缺条关联分析模型,挖掘特征库中的烟箱缺条特征,获取烟箱缺条的极大频繁项集,输入所有待检测的烟箱,通过特征匹配和关联规则获取智能检测结果,完成烟箱缺条智能检测。测试结果表明该方法具备较好的数据挖掘性能和效率,在支持度和置信度均为最小值时,可准确完成烟箱缺条特征匹配,在不同噪声条件下,均具备较好检测结果,应用性较好。
Aiming at the problems of low detection accuracy and poor noise resistance in the detection of tobacco box missing strip,the application of association rule mining method in the intelligent detection of tobacco box missing strips on the conveyor line is studied.Based on the concept of association rules mining,combined with the concept of feature matching degree,the paper obtains all the features of the tobacco box in the conveying line,and then constructs the tobacco box missing strip feature library.Using the tobacco box missing strip association analysis model,the paper mines the tobacco box missing strip features in the feature library,obtains the maximum frequent item set of tobacco box missing strip,inputs all the tobacco boxes to be detected,and obtains the intelligent detection results through feature matching and association rules to complete the intelligent detection of cigarette box missing strips.The test results show that the method has good data mining performance and efficiency.When the support and confidence are the minimum,it can accurately complete the tobacco box missing feature matching,under different noise conditions,it has good detection results and good applicability.
作者
宋煜
江志凌
刘艳超
SONG Yu;JIANG Zhiling;LIU Yanchao(Information Management Office,Wuhan Cigarette Factory,Wuhan 430040,China)
出处
《微型电脑应用》
2023年第4期202-204,208,共4页
Microcomputer Applications
关键词
关联规则
挖掘方法
输送线
烟箱缺条
极大频繁项集
association rule
mining method
conveyor line
cigarette box missing strip
maximal frequent item set
