基于小波变换模极大值识别竹节纱外观参数的研究

Research on Appearance Parameters Identification for Slub Yarn Based on Wavelet Transform Modulus Maxima

通过对竹节纱外观信号的观察,根据小波变换模极大值对信号奇异性检测的原理,识别出竹节纱竹节部分并实现竹节与基纱分界点的定位。为了提高定位的准确性,通过交替投影算法实现对竹节纱的信号重建,同时分析了纱线条干不匀和棉结对竹节识别的影响,设定阈值在算法设计中消除两者对竹节识别的干扰,最后设计出竹节长度、竹节间距、竹节倍率和基纱细度这4个外观参数的计算公式,为新型竹节纱检测仪器的开发提供技术参考。

Through the observation of the appearance signal of slub yarn and according to the principle of signal singularity detection based on wavelet transform modulus maxima,the slub part of slub yarn was identified,and the positioning of the slub and base yarn demarcation point was achieved.In order to improve the accuracy of positioning,signal reconstruction for slub yarn was achieved through alternating projection algorithm.Meanwhile,the influence of cotton yarn non-evenness and twinning on slub identification was analyzed.A threshold was set to eliminate the interference of both on slub section identification in algorithm design.Finally,the computational formula of four appearance parameters（slub length,slub spacing,slub rate and base yarn fineness）was designed to offer some references for development of new slub yarn detection instruments.