A Dynamically Compensated Center of Gravity Method for Peak Extraction in Confocal Microscopy

Abstract

The invention discloses a method for dynamically compensating the center of gravity for confocal microscope peak extraction, wherein the above method mainly includes the following steps: obtaining a sampling sequence, sampling to construct a standard effective sampling sequence, and further sampling to construct two types of offsets. Effective sampling sequence, calculate the extracted peak value according to the center of gravity method, calculate the quantitative relationship between the systematic error and the offset in six blocks, and subtract the system error of the corresponding block from the error of each block obtained by the center of gravity method to obtain the biased Dynamic compensation results under displacement. The method for dynamically compensating the center of gravity for confocal microscope peak extraction realized according to the invention greatly reduces expected error and uncertainty level, and significantly improves calculation efficiency and real-time algorithm accuracy.

Description

technical field [0001] The invention belongs to the field of optical precision measurement, in particular to a method for dynamically compensating the center of gravity for confocal microscope peak extraction. Background technique [0002] In confocal microscopy measurement, the premise of accurate acquisition of height information is that the peak position extracted from the sampled axial response curve remains fixed relative to the optical system (objective lens). However, there must be various noises in the sampling signal, such as detector noise, scanning positioning error and so on. Among them, the maximum value method in an algorithm in the prior art directly selects the height corresponding to the point of maximum light intensity as the peak value. This method is extremely susceptible to noise, and can only distinguish the change of a sampling interval; wherein the prior art Another algorithm, the center of gravity method, is a fast algorithm with strong robustness t...

AI Technical Summary

Problems solved by technology

However, there must be various noises in the sampling signal, such as detector noise, scanning positioning error, etc.

Among them, the maximum value method in an algorithm in the prior art directly selects the height corresponding to the point of maximum light intensity as the peak value. This method is extremely susceptible to noise, and can only distinguish the change of a sampling interval; wherein the prior art Another algorithm, the center of gravity method, is a fast algorithm with strong robustness to noise, published in "Measurement Science and Technology" in 2015 in "Sinc2 fitting for height extraction in confocal scanning" document 1 (document publication information: Tan J, Liu C, Liu J, et al.Sinc2 fitting for height extraction in confocal scanning[J].Measurement Science&Technology,2016,27(2):025006.) pointed out that fitting methods such as parabolic fitting, Gaussian fitting, Sin 2 The C fitting method has a smaller algorithm uncertainty, that is, the noise robustness is better than that of the center of gravity method

However, the "Signal evaluation for high-speed confocal measurements" document 2 published in "Applied Optics" in 2002 (document publication information: Ruprecht A K, Tiziani H J, Wiesendanger T F. Signal evaluation for high-speed confocal measurements[J].Applied Optics,2002,41(35):7410.) points out that in actual measurement, if the height of the measurement object changes linearly, the height of the measurement object is unknown before measurement, so it may be an arbitrary result. There will be a linearly increasing offset in the response curve, which will cause a systematic error in obtaining the peak position using the center of gravity method

Although the research literature in 2002 pointed out the existence of the offset and proposed a correction method, the correction of this method is strictly based on the relationship between the sampling interval and the half-height width, and the quantitative relationship between the two is difficult to be accurate in actual measurement. It is not easy to obtain the interrelationship between the two on which the corrective method depends, which reduces the effectiveness of the corrective method

However, the influence of the coupling effect of the two on peak extraction has not been solved and has not been studied. However, the influence of the coupling effect of the two on peak extraction is an important prerequisite for accurate acquisition of height information.

Images