Research On The Key Technology Of Quantitative Phase Imaging Based On Digital Holography

Digital holographic quantitative phase imaging is a non-contact,high-resolution,label-free,fast reconstruction,low-cost 3D imaging technology.It has broad application prospects in biomedicine,industrial manufacturing,material science and other fields.In phase reconstruction,the accuracy of imaging directly affects the accuracy of recovering the true shape of the object,while obtaining phase information in ultrafast dynamic scenes is also a key factor in the further development of the technique.However,the speckle noise generated by the laser will affect the phase accuracy of digital holographic imaging.The research on speckle noise suppression methods carried out in this thesis aims to address this problem.In addition,the field of ultrafast digital holography quantitative phase imaging is also explored.The main contents are as follows:(1)The principles of digital holographic recording and numerical reconstruction processes,as well as the origin and formation processes of speckle noise,are deduced.And the influence of speckle noise on quantitative phase imaging is analyzed.Build a digital holographic optical path and use a detector to capture holograms of phase-type objects.(2)For simpler samples with a single-phase profile,a discrete cosine filtering method based on the sine-cosine decomposition is proposed to suppress the speckle noise in the wrapped phase.In which,the wrapped phase with severe jump phenomenon is transformed into a continuous parameter by using the sine-cosine transform.Then combined with the properties of the discrete cosine transform to suppress the speckle noise.The feasibility of the algorithm is verified by reconstructing the true phase distribution of a simple object.Compared with existing denoising methods,this method is simple and fast with higher accuracy in phase reconstruction.(3)Considering that the discrete cosine filtering method based on the sine-cosine decomposition may cause the loss of high-frequency information,which leads to the loss of information about the details of the reconstructed phase distribution.Therefore,an improved P-M equation based on empirical mode decomposition is proposed to suppress the speckle noise.The high-frequency information of the hologram is highlighted using empirical mode decomposition,and edge detection is performed using the Canny edge detection operator.The high-precision detection results are used as a guide for diffusion denoising to achieve better denoising effects.The method is compared with the discrete cosine filtering method based on the sine-cosine decomposition and existing denoising methods.The contrast,structural similarity and edge preservation index have increased by 23.0%,5.9% and 13.9%,respectively.The speckle suppression index is reduced by a maximum of 14.9%.The index calculation results and phase cross-section curves show that the method can retain more details to obtain more accurate phase information by effectively suppressing the speckle noise.(4)Combining the existing digital holography technology with compressed ultrafast photography to reconstruct the phase information of ultrafast dynamic scenes.The feasibility of the technique is verified by simulation,which provides research ideas for the field of ultrafast digital holography quantitative phase imaging.Among them,the traditional compressed ultrafast reconstruction algorithm is replaced by a plug-and-play algorithm based on the alternating direction multiplier method.The highest peak signal-to-noise ratio is26.6057,and the highest structural similarity is 0.9779.A multi-frame dense hologram fringe scene is recovered with high quality from a two-dimensional superimposed compressed hologram at each time point,which provides a guarantee for high-precision phase reconstruction.In this thesis,we conducted a related study on digital holographic quantitative phase imaging technology and proposed a method to suppress the speckle noise,which makes the phase reconstruction accuracy effectively improved.We also made a preliminary exploration of the field of ultrafast digital holographic quantitative phase imaging,which helps promote the further development of this technology.