Research On Microscopic Multispectral Imaging Technology Based On Fourier Ptychographic

One of the goals in microscopic imaging is to obtain multidimensional and high-resolution information of the sample being measured,including spatial,spectral,field-of-view,and temporal dimensions.However,current microspectral imaging techniques are unable to simultaneously achieve high spatial resolution,wide field-of-view,and high spectral resolution.Fourier ptychographic microscopy(FPM)is a novel microscopic imaging technique that can achieve a wide field-of-view while maintaining high spatial resolution.However,conventional FPM systems typically use narrow-band LEDs as illumination sources,which only capture sample information within a specific spectral band and result in the loss of spectral information.To address this limitation,this paper introduces the imaging link and recovery algorithm of conventional FPM,analyzes the research on spectral multiplexing in FPM,and proposes an optimization method and an image colorization application based on this approach.To achieve high spatial resolution,high spectral resolution,and wide field-of-view microscopic imaging,this paper proposes a multispectral imaging method based on FPM,which includes two aspects:(1)The paper studies spectral imaging in red,green,and blue wavelengths and performs color image synthesis using FPM.The imaging setup and recovery algorithms of conventional FPM imaging methods based on red,green,and blue wavelengths,as well as FPM imaging methods based on spectral multiplexing,are studied.The advantages and disadvantages of these two imaging methods are discussed,and corresponding optimization schemes areproposed.Through simulation and real-life experiments,it is demonstrated that both imaging methods can recover discrete spectral images with high spatial resolution and wide field-of-view,and the discrete spectral images can be synthesized to obtain color images.(2)The paper mainly focuses on multispectral imaging based on FPM and proposes an optimized FPM setup using a dual-camera system with a full-band halogen lamp and narrow-band filter illumination.In this setup,different wavelength illumination sources are obtained by replacing filters of different central wavelengths.A series of low spatial resolution images of different wavelengths are acquired by grayscale cameras,and then discrete spectral images are recovered by recovery algorithms.The mapping relationship between discrete spectral images and color images acquired by color cameras is analyzed for high spectral resolution reconstruction,and images of high spatial resolution,high spectral resolution,and wide field-of-view are reconstructed.The validation of the setup and the hyperspectral reconstruction algorithm is carried out by simulation experiments and real-world experiments,respectively.The experimental results show that the image data cube reconstructed by the proposed method has high spatial resolution,wide field-of-view,and a spectral resolution of 5nm.