Accurate and efficient height extraction in chromatic confocal microscopy using corrected fitting of the differential signal
Accurate and efficient height extraction in chromatic confocal microscopy using corrected fitting of the differential signal
复制标题
使用差分信号的校正拟合在彩色共焦显微镜中准确高效地提取高度
DOI：
10.1016/j.precisioneng.2019.02.001
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Jiang Xiangqian
中科院分区：
文献类型：
--
作者：
Chen Cheng;Yang Wenjun;Wang Jian;Lu Wenlong;Liu Xiaojun;Jiang Xiangqian
准确高效的峰值提取算法对于实时彩色共焦测量至关重要。然而，当前最先进的算法无法同时满足高提取精度和计算效率的要求。在本文中，我们提出了一种具有良好精度和高计算效率的修正拟合微分算法（CFDA）。首先，CFDA通过差分信号的常规线性拟合来提取初始峰值波长，该差分信号是从光谱轴向强度信号导出的。其次，拟合误差通过广义微分方程来表征。最后，通过用一阶近似求解该方程来估计和校正误差。与高精度高斯拟合算法相比，我们的模拟和实验表明，CFDA 可以在保持相当的峰值提取精度的同时，实现计算效率约一百倍的提升。
Accurate and efficient peak extraction algorithm is essential in real-time chromatic confocal measurement. However, current state-of-art algorithms cannot simultaneously satisfy the requirement of high extraction accuracy and computational efficiency. In this paper, we propose a corrected fitting differential algorithm (CFDA) with good-accuracy and high computational efficiency. Firstly, the CFDA is conducted to extract the initial peak wavelength with the conventional linear fitting of the differential signal, which is derived from the spectral axial intensity signal. Secondly, the fitting error is characterized via a generalized differential equation. Finally, the error is estimated and corrected by solving this equation with a first-order approximation. Compared to the high-accuracy Gaussian fitting algorithm, our simulations and experiments demonstrate that the CFDA can achieve approximately a hundred times improvement in computing efficiency while maintaining a comparable peak extraction accuracy.