Research On Image Reconstruction Based On Integral Field Unit Spectrometer

The field of astronomy utilizes celestial spectra to obtain rich celestial information,thus the development of astronomical observation technology at home and abroad is in full swing.In order to better achieve celestial observation,some telescopes are equipped with spectral image acquisition equipment,such as the Fiber Optic Array Solar Optical Telescope(FASOT).This telescope collects the light emitted from a certain area on the solar surface and uses a microlens array for pupil division.The two-dimensional field of view is sampled using an integrated field unit(IFU),and then inputted into the spectrometer through optical fibers to obtain three-dimensional information of the spectrum(spatial x,y,and spectral domainsλ)。At present,image reconstruction technology can be applied in fields such as medical imaging,remote sensing,video processing,deep learning,and art entertainment to improve image quality,remove noise,restore missing information,and improve resolution.In the field of astronomical spectral image reconstruction,the reconstruction method of microlens array integrated field of view fiber spectrometer is currently rarely studied.This article aims to deeply explore and supplement the image reconstruction methods in this field in detail,providing a new approach for spectral image reconstruction methods in the field of astronomy.This thesis proposes an image reconstruction method for a micro lens array integrated field of view fiber spectrometer.Firstly,contour fitting is used to approximate the flow rate values of spatial sampling points,and a functional expression describing the fiber flow rate is generated to obtain the final fiber flow rate value,thereby obtaining one-dimensional spectral energy.Wavelength calibration is performed by calibrating the lamp spectrum;Then,by using the shape of the fiber optic array end,which is a regular hexagonal staggered arrangement,the contour of the original celestial body image is reconstructed.Through the one-to-one correspondence between the IFU fiber optic array end and the slit end fiber,reverse fitting is performed to reconstruct the fixed wavelength or grayscale image of the original celestial body at a fixed wavelength,completing the band reconstruction;Finally,high-resolution images of primitive celestial bodies are obtained through super-resolution reconstruction methods.In this thesis,SSIM,PSNR and other image quality evaluation indexes are used to evaluate different images obtained by the image reconstruction algorithm.Among them,there are 11×11,64×64 optical fiber target holes of the optical fiber array end image and their super resolution reconstruction images.Through a series of correlation experiments,the final experimental results show that,this algorithm is a feasible spectral reconstruction algorithm,which can realize image reconstruction,and has certain advantages in resolution.This thesis also extends the application of temperature analysis of celestial radiation components.The above methods are integrated into the system,and the multi-function display is realized,including single-fiber one-dimensional spectrum display,multi-fiber one-dimensional spectrum display,fixed component temperature analysis,image reconstruction.The research concept proposed in this thesis further explores the reconstruction of spectral images,and the spectral image reconstruction method proposed in this thesis can currently be applied to astronomical fields such as neighboring galaxies.In addition,the method proposed in this thesis can also be applied to solar multi band imaging observations,and is currently an important means of measuring the sun.Specifically,this study can analyze phenomena of solar atmospheric activity,such as multi band analysis of the fine structure of solar flares,chromospheric features of small-scale magnetic phenomena,and the photosphere foot points of the solar corona ring,which are of great significance.Therefore,the research results of this thesis have broad implications.