小波变换轮廓术抑制CCD非线性的分析  

Analyzing wavelet transform profilometry in the restraining CCD nonlinear characteristic

作　　者：苏轲[1] 陈文静[1] 

机构地区：[1]四川大学电子信息学院光电系,成都610065

出　　处：《光学技术》2009年第1期37-40,44,共5页Optical Technique

基　　金：国家自然科学基金资助项目(60677028)

摘　　要：对比研究了CCD非线性对小波变换轮廓术和傅里叶变换轮廓术的影响,并从信号频域角度分析推导出考虑CCD非线性时,变形条纹的小波变换的频谱描述形式,得到了"脊"处小波系数的解析表达式。处理由CCD非线性引起的非完善的条纹图时,采用小波变换轮廓术提取相位,实质是采用最佳的加权滤波窗口,这样能减弱CCD非线性引起的频谱混叠对测量的影响,可以得到比傅里叶变换轮廓术更稳定的恢复效果。计算机模拟验证了此结论的正确性。From the point of the nonlinear characteristic of a CCD camera, it is focused on studying the error distributions caused by the nonlinear in wavelet transform profilometry and Fourier transform profilometry. Then the analytical frequency-domain description of the wavelet coefficient is deduced when CCD nonlinear characteristic is considered. The error from the frequency overlapping caused by CCD nonlinear chrematistic can be eliminated to a low extent because of an adaptive optimized filtering operation when wavelet transform profilometry is applied. Therefore the measurement accuracy of WTP is higher than FTP in that case. This analysis is confirmed on the computer simulations and experiments.

关 键 词：信息光学 三维传感 小波变换轮廓术 傅里叶变换轮廓术 条纹分析 

分 类 号：TN247[电子电信—物理电子学]