基于模糊专家系统的印刷品缺陷分析方法  

Printing defect analysis method based on fuzzy expert system

作　　者：文为[1,2] 孙晓刚[1,2] 张亮[1] 

机构地区：[1]中国科学院成都计算机应用研究所,成都610041 [2]中国科学院大学,北京100049

出　　处：《计算机应用》2014年第A01期228-231,共4页journal of Computer Applications

摘　　要：针对印刷品缺陷分析中存在的模糊性和不确定性的问题,提出了一种基于模糊专家系统的印刷品缺陷分析方法。该方法根据检测出的印刷品缺陷、生产线的各种工艺参数及专家对缺陷出现原因的知识建立专家知识库,完成了基于产生式规则的模糊知识表示和模糊表示下知识数据库的建立与推理机的实现。该技术充分利用了已有的专家经验和知识,解决了缺陷分析中存在的模糊性和不确定性问题,能准确推测出印刷品缺陷的产生工序及判废结果。实验结果表明,使用基于该方法的印刷品缺陷分析系统进行缺陷分析时,平均准确率为84%,满足现阶段印刷工业分析要求。In order to solve the fuzziness and uncertainty problems in the defect analysis, this paper proposed a printing defect analysis technology based on fuzzy expert system. This technology established expert knowledge based on the detected print defects, the various process parameters of the production line and the knowledge of the cause of the defect appeared, completed the production nile-based fuzzy knowledge representation and the construction of knowledge databases based one fuzzy representation, then implemented inference engine. The technique takes advantage of the existing expert knowledge to solve the fuzziness and uncertainty problems in the defect analysis. The proposed method, can accurately infer the generation processes of print defects and rejection results. The experimental results show that the average accuracy rate is 84%, which meets the requirements of the printing industry in the present stage .

关 键 词：印刷品 表面缺陷 模糊专家系统 缺陷分析 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]