Solar Radio Image Processing Based On Multi-scale CLEAN Algorithm

The ultra-wide spectrum radio heliostat(MUSER)located at the location of Ming Antu in China is a special radio interference array for the sun,which can simultaneously produce high-time,high-space and high-frequency resolution solar images.However,according to the comprehensive aperture imaging principle of the heliostat,its sampling points in the UV plane are limited,so the obtained sun image is not real,and it will include many false components.Therefore,it is necessary to use a specific image processing method to restore the observed sun image to obtain a real image.Solar radio imaging relies on a deconvolution algorithm to cancel the sparse sampling of the Fourier plane.The most widely used radio synthesis deconvolution method is H(?)gbom's CLEAN(cleaning)algorithm,which is very suitable for point source collections,but it does not work very well for spread source objects.In this paper,we use the multi-scale CLEAN algorithm as the deconvolution method in MUSER solar radio imaging,which is applicable to both point and spread sources.This paper compares and improves various parameters to obtain the optimal results.The main research work is as follows:1.Discusses the development of solar radio observation instruments at home and abroad,highlights the built-in Ming'antu ultra-wide spectrum radio heliostat(MUSER)and its parameters,and summarizes the domestic and foreign cleansing of the image of the heliostat Research status.Based on the working principle of the interferometer,the comprehensive aperture imaging principle of the heliostat is analyzed,and then the basic CLEAN algorithm ideas are introduced,and some existing CLEAN expansion algorithms are described.2.Through specific analysis and research on H(?)gbom CLEAN algorithm,multi-resolution CLEAN algorithm and traditional multi-scale CLEAN algorithm,and using them to clean MUSER solar integrated aperture radio imaging.Comparative analysis of the implementation results of these algorithms in MUSER imaging proves the advantages of the multi-scale CLEAN algorithm in solar image processing.3.Based on Cornwell's traditional multi-scale CLEAN algorithm,corresponding improvements were made.AIA(Atmospheric Imaging Assembly)image was used to conduct simulation experiments to study the multi-scale CLEAN algorithm on factors such as scale selection and iteration times,and adopt objective evaluation indicators.Evaluate the restored image.Then,the improved multi-scale CLEAN algorithm is used for image processing of MUSER solar data,and compared with the traditional multi-scale CLEAN implementation results.The optimized multi-scale CLEAN algorithm in this paper shows in the results of MUSER imaging processing that it has a good processing effect on point sources and spread sources,and is more effectively suitable for the processing of solar images,with better solar radio imaging The performance has a good reference value in the research and development of radio astronomy topics.