摘要
在无损检测信号处理和特征构造的基础上,将语音信号处理领域的隐玛尔柯夫模型(HMM)引入无损检测缺陷模式识别中,辅以计算机控制系统,实现了航空材料内部缺陷识别的定量化和智能化,这对确保航空材料的质量和安全使用具有重要的实际意义.
In this paper, a method of defect type recognition based on Hidden Markov Model(HMM), which is initially employed in the field of speech signal processing, is presented for non-destructive testing (NDT), where the signal is processed and the characteristic is constructed. It realizes quantitative and intelligent recognition of defect patterns in aviation material with the help of a parameter analysis method based on fuzzy degree of membership and the Baum-Welch training algorithm. It has practical significance for ensuring the quality and safe use of aviation material.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2002年第4期489-493,共5页
Pattern Recognition and Artificial Intelligence
关键词
隐玛尔柯夫模型
无损检测
缺陷
模式识别
铸钢缺陷
Non-Destructive Testing(NDT), Hidden Markov Model(HMM), Pattern Recognition
