Study On Fluorescence Microscopy Image Reconstruction For Organic Thin-film Growth

This paper is based on a real-time in situ fluorescence microscopy system for investigating the organic thin film and firstly introduces this system.Image noise and variable refractive index optical path are proposed on the basis of detailed analysis of main factors affecting the image quality of thin film growth.For the influence of image nosie,the choice of threshold,threshold function and some key parameters are analyzed in detail based on wavelet denoising.And combined with the noise characteristics in the actual CCD image acquisition process,a new threshold function——Power threshold function is presented.This function is the spread of soft and hard threshold function,which overcomes the shortcomings of the hard threshold function which is discontinuous and the soft threshold function which has deviation.MATLAB simulation results show that the threshold function is better than soft and hard threshold function on the PSNR and MSE,which has better de-noising effect.In addition,transplanted this algorithm to auto-focus microscopic image acquisition system,image definition is greatly enhanced which is beneficial to capture clear imaging surface quickly and accurately.Furthermore,this paper also discusses the imaging impact of variable refractive index optical path after adding the observation window.Based on geometric optics knowledge,theoretically analyzes the change in the optical paths which result by different transparent medium thickness and refractive index,and obtains important conclusion: the offset of focus position exhibits linear change not only along with the thickness of transparent medium but also with the reciprocal of its refractive index.Finally,the correctness of the above conclusion is verified by ZEMAX optical simulation software.