Algorithm Research On Reconstruction Of Depth For Microscopic Specimen Based On 3D-Furier Ptychography Microscopy

Traditional microscopic observation always concerntrates on 2-dimensional distribution of biological tissure,however,with the fast development of pathology and biomedical engineering,more and more attention has been paid to the distribution on 3 dimension.The technology,which has the ability to observe the specimen in lateral and axial direction,is termed 3D microscopic imaging technology.By reconstructing the 3D information of biological specimen,3D microscopic imaging techonology can help the doctors to determine the position and morphogogy of foci,having important and far-reaching significance on the morphological determination of tissure as well as the patholotical diagnosis.With regard to FOV and resolution as the inherent trade-offs of optical system,point scaning or aperture stich method is used for obtaining large field of view.To capture a sequence of images along z axis,a moving part is always neccesary for microscopic system.But,there exist some shortcomings:(1)Since point scanning is time consuming,it’s not praticle to observe specimen in real-time;(2)the fast moving part of the microscope stage(or objective)during the refocusing can perturb the specimen,rendering unprecise measurement;(3)large amount of images along z axis do lead to large data to be processed,and enhance the computational burden of the algorithm.To solve above problems,in this paper,a 3D microscopic reconstruction technology for depth is proposed,which conquers the inherent trade-off between FOV and resolution.The corresponding low-cost system has the ability to image microscopic specimen with high resolution and large field of view,providing the possibility for fast,even real-time 3D reconstruction for depth.The main contents of this paper consists of following parts:Firstly,the present situation of 3D imaging technology is introduced here,which analyzes the system and algorithm in detail,and the imaging method in this paper is proposed subsequently;Subsequently,detailed illustration of the theoretical basis and system structure are made to derivate the operation of lateral resolution and axial resolution,and offers the theoretical basis for the improvement of algorithm;in re the shortcoming of the present alogorithm,there are two modifications:selection of optimization routine and adaptive selection of illumination based on energy in Fourier domain;on the basis of characteritics of the improved algorithm,it is better to adopt non-reference image quality method to evaluate the quality of our algorithm results;moreover,based on the alogorithm proposed here,we carried out the simulation and experiments of our algorithm,the results of which are basicaly in conformity with the theory,the resolution being over diffraction limit;in the last chapter,work summay and future development are made in this paper.