Research On ONSET High-resolution Image Reconstruction And Mosaic Method

Observing the sun and obtaining corresponding observation images are common methods for studying solar physics.However,because of the effects of atmospheric turbulence,the images obtained by ground-based astronomical telescopes are subject to random distortion,which affects the resolution and clarity of the images.In order to solve the impact of atmospheric turbulence,one of the most effective methods for obtaining high-resolution images from high-resolution image reconstruction techniques that began in the 1970 s.The Optical and Near Infrared Solar Outburst Detection Telescope(ONSET)is a small aperture multi-channel solar special telescope that has been working at the Chengjiang Solar Observatory in Yunnan for many years.Based on systematic analysis of the current mainstream high-resolution image reconstruction methods,this paper systematically studies the displacement superposition method.It is believed that this method can meet the requirements of real-time rapid image reconstruction due to the relatively small amount of calculations.ONSET realizes high-resolution solar observations.An effective method of data processing.Based on previous work,this paper focuses on the two existing problems in ONSET reconstruction methods,namely,image stitching,image deconvolution,and image enhancement methods.In image stitching,this paper discusses two methods,namely,The usability of SIFT affine transformation method and wavelet fusion method is used in image deconvolution.This paper discusses the method of deconvolution of Wiener filter.In image enhancement,this paper discusses the image enhancement method based on log algorithm.The specific work of this article is as follows:(1)To investigate current research status and reconstruction methods of high-resolution image reconstruction,and to study some existing problems in the existing ONSET telescope image reconstruction methods.(2)The SIFT affine transformation method and wavelet fusion method are studied in the splicing part of the existing ONSET telescope image reconstruction method.The 4250 ? band image of ONSET telescope is stitched using SIFT affine transformation method and wavelet fusion method.The methods have all been improved,but for the splicing part of the H?-bandimage of the ONSET telescope,the image splice by the wavelet fusion method works well,and the SIFT affine transformation method cannot complete the matching because the matching sub-blocks are too few.Image stitching.(3)Some improvements have been made to the existing ONSET telescope image reconstruction methods,focusing on image deconvolution and image enhancement after image splicing.The Wiener filtering method is used to deconvolute the spliced image.The log algorithm is used to enhance the deconvoluted image.The quality of the image is improved.The improved image reconstruction method improves the stitching of the image stitching on the stitching,which makes the image over-natural and rich in details.The comparison through multiple sets of experiments shows that the method is stable,and the image with unsatisfactory image quality can also be improved.Rebuilding an ideal image effectively.