Research On Multi-scale Velocity Field Measurement Methods For High-resolution Images Of The Sun

There are a large number of different scales,different frequencies and different speeds of solar activity in the solar images.Such as large-scale activities(sunspots,flares,filament etc)and small-scale activities(microflare,minifilament etc).It is very difficult to measure the particular target velocity field when the solar activity is mixed at different scales.This paper aims to calculate the velocity field with different spatial scales by using the idea of multi-scale and multi-resolution combined with optical flow technique.Therefore,we can get the movement trend of the solar activity with different scales,and understand the motion law of the solar features from the local to the whole.In view of the above problems,the following studies are carried out in this paper:Firstly,in this paper,the traditional single scale velocity field measurement method is analyzed,and then the local correlation tracking algorithm is improved.The main content of this paper is the search strategy of thickness matching,and the correlation coefficient is used to weight the local area.Local correlation tracking and weighted local correlation tracking are compared.Secondly,in order to measure the velocity field of different scales of solar activity,the weighted local correlation tracking algorithm and multi-resolution hierarchical refinement which combines to realize multi-scale weighted local correlation tracking algorithm to measure the mixture velocity.Thirdly,in order to verify the accuracy of the multi-scale weighted local correlation tracking algorithm,the error analysis is carried out first by simulation experiment.The results show that the improved measurement error of the multi-scale weighted local correlation tracking algorithm is 1%.Then,the velocity field of three real data sets are measured.Not only for the sunspreads and penumbra of the speed field to accurately measure,and in which the two data sets,also found in the sun within the shadow of the movement,which is the traditional single-scale velocity field measurement method can not be get.Fourthly,the important parameters of multi-scale weighted local correlation tracking algorithm are discussed,including the number of layers and the size of the window in Pyramid.This method can be used to analyze the change of adjacent time and consider the change of image content from the point of view of the whole sequence image.By constructing a Pyramid image,we can quantitatively analyzed the velocity field at different resolutions,obtained the velocity field at different spatial scales,showing the motion trend of solar activity characteristics at various scales.