Spectral Signal Asymmetry Analysis and Encoding of Chromatic Confocal Sensors

The chromatic confocal technique is a precise, noncontact, and nondestructive optical measurement method. The measurement accuracy of the chromatic confocal sensor is heavily reliant on the precision of the signal evaluation algorithm. However, existing algorithms face limitations due to the asymmetry of the spectral profile and the presence of random noise. In this study, a chromatic confocal signal model based on Fourier optics theory is established to analyze the influence of various factors, including nonconformity in light source response, optical system aberration, and surface features of the measured object, on the profile shape of the spectral signal. To quantify the distortion of the signal profile, asymmetric parameters are introduced. In addition, we propose a novel spectral coding method that combines a variable threshold function with spectral fitting using a hybrid model, namely the exponentially corrected Gaussian and asymmetric Voigt function (EMG-AVF). This method demonstrates reduced measurement uncertainty compared to conventional line shapes. To verify the accuracy of our proposed method, we conducted experiments involving step height measurements and tilted plane mirror measurements.