提高莫尔条纹正切法细分精度的改进算法  

New Algorithms for Improving Tangent Method Subdivision Precision of Moiré Fringe Signal

作　　者：吕孟军[1] 张纯良[1] 游有鹏[2] 徐峰[1] 

机构地区：[1]空军第一航空学院航空机械工程系,信阳464000 [2]南京航空航天大学机电学院,南京210016

出　　处：《纳米技术与精密工程》2011年第3期194-198,共5页Nanotechnology and Precision Engineering

基　　金：江苏省精密与微细制造技术重点实验室开放基金资助项目(20060306)

摘　　要：为提高莫尔条纹信号正切法细分的精度和倍数,应用改进算法对原始信号进行数据预处理.通过神经网络自适应算法在莫尔条纹信号降噪方面的移植,结合滤波步长的实时调整,有效滤除了信号的宽频段有色噪声干扰;采用了基于滞后相位补偿思想的相位调整算法,通过分段处理,提高了原始正余弦信号的相位正交性.对算法处理后的莫尔条纹信号进行正切法细分验证,实验结果表明当细分倍数为1 024时,可确保细分误差小于1个细分当量,即0.618″,若不考虑信号的非正弦性影响,细分精度提高了10倍以上.In order to enhance the accuracy and multiple of Moiré fringe signal＇s tangent subdivision,the improved algorithms were applied to pretreating the original signal data.Combined with real-time adjustment of filtering step,the adaptive neural network algorithm was used to restrain the noise of Moiré fringe signal,so that full band colored noise was effectively filtered.A new phase adjustment algorithm based on lag phase compensation and section treatment was used to improve phase orthogonality of sine and cosine signals.Experiments of the tangent subdivision of Moiré fringe signal treated with the proposed algorithm were conducted and results showed that when subdivision multiple was 1 024,subdivision error was less than a subdivision equivalent 0.618″ and subdivision accuracy was increased by more than 10 times in case there was no signal distortion.

关 键 词：莫尔条纹 正切法 细分精度 算法 

分 类 号：TN911.7[电子电信—通信与信息系统] TP273[电子电信—信息与通信工程]