Research Of Super-Resolution Microscopic Imaging Technology Based On Multiple Signal Classification Algorith

In the field of living cell imaging,optical microscopy stands out among many methods because of its nondestructive and non-contact imaging properties.However,the existence of optical diffraction limit limits its imaging resolution.In order to realize the observation of smaller cell and subcellular biological samples,it is of great significance to improve the imaging resolution of optical microscopy.Confocal microscopy has unique advantages in axial tomography,but its transverse resolution is only 1.4 times of that of wide-field microscopy.However,there are many limitations in the existing improved methods.In view of the problem of low horizontal resolution of confocal microscopy technology,this dissertation from the perspective of spatial spectrum estimation theory,proposes a super resolution microscopy imaging method based on multiple signal classification algorithm,the multiple signal classification algorithm(MUSICALS)combined with the point lighting of confocal microscopy technology.Through singular value decomposition of the original spot image sequence collected by the ordinary superresolution scanning system,the spatial spectrum estimation is carried out by using the orthogonal signal subspace and noise subspace,and then the high resolution imaging results are obtained.Compared with super-resolution techniques such as structural detection,ISM and SOFI,which can be combined with confocal scanning microscopy,this method has the advantages of high imaging resolution and strong noise suppression ability.However,this method has the following disadvantages: low SNR and high order information are easily submerged in the reconstruction results,and the imaging speed needs to be improved.Aiming at the problem of submerging low SNR and high order information,the reasons are analyzed and the improved method is studied.Finally,the effectiveness of the improved method is verified by experiments.In terms of imaging speed,the influence of lighting spot quality on imaging results is simulated and analyzed in this paper,which lays the foundation for further improving the image acquisition speed to make up for the disadvantage of this method in imaging speed.The main works are as follows:(1)Firstly,studying the multiple signal classification algorithm,and establish the multiple signal classification algorithm based on super resolution theory model of microscopic imaging technology,the production,light flashing sample delivery process,data collection,processing and image reconstruction simulation,verify that the method is2.8 times the ordinary confocal microscope.(2)Investigate the imaging characteristics of the multiple signal classification algorithm,the simulation analysis of the threshold,index factor,proportion of pixel size,window size,the influence on the result of parameter selection,to drown of low signalto-noise ratio and high order information problem,the research of soft threshold and threshold weighting method,and analyzes its for submerged higher-order low fluorescence characteristic vector and brightness the recovery effect of the sample.(3)The effect of aberration of lighting system on imaging results is analyzed by simulation.In the confocal system,moving beam scanning or array scanning can significantly improve the imaging efficiency of the system.However,due to the introduction of mechanical scanning mode and array elements,the distortion and position deviation of the illumination spots are easy to be caused,thus reducing the imaging resolution of the system.Therefore,the influence of illumination spot aberration and light intensity stability on confocal microscopic imaging results based on multiple signal classification algorithm will be simulated and analyzed in this paper to provide a reference for further improving the image acquisition speed of this method.(4)Finally,a super resolution microscopic imaging verification experiment based on multiple signal classification algorithm is carried out.Designed and built based on the multiple signal classification algorithm for super resolution microscopy imaging system experimental platform,the original image of patch lighting,data processing and image reconstruction,validation based on multiple signal classification algorithm for super resolution microscopic imaging characteristics of high resolution,validation of soft threshold and threshold weighting method of multiple signal classification algorithm imaging results of contrast is optimized.