Sunspot Low Contrast Feature Recognition

The sun is a star that located in the center of the solar system,it has a great impact on the Earth's creatures.The lowest solar layer is the solar photosphere,but its brightness distribution is not uniform.In the solar photosphere layer,the quiet areas contain granulation,and the activity areas contain sunspots,spots and so on.We study the low contrast characteristics in the sunspots-the umbral dots.There are some small isolated brightenings that we can observe in sunspot umbrae,called umbral dots.They are connective phenomena in sunspot that almost exist in every umbra.The study of umbral dots for understanding connective motions,the plasma interactions with strong magnetic fields,and analyzing the formation mechanisms of sunspots is very important.And so far,the existed recognition methods are mostly based on the intensity and gradient methods.However,because of the characteristics of low intensities,small gradients and unclear boundaries of umbral dots,the methods that based on intensity and gradient are difficult to identify umbral dots accurately.Consequently,in order to improve the accuracy of umbral dots identification,we propose a new method that based on phase congruency to extract and identify umbral dots.In this paper,we select two high resolution image sequences to introduce the proposed method,including a data set taken from the Hinode/Solar Optical Telescope on 2 March 2007,another data set obtained with New Vacuum Solar Telescope.The procedures of identification are as follows:(1)firstly,we detect the umbra-penumbra boundaries of sunspots by using the existed technique,In order to better show the internal structure of sunspots,we adjusted the gray scale of the original image,we extract the umbral regions for the latter experiment;(2)secondly,we use phase congruency technique to identify umbral dots.At last,we identify the statistical properties of umbral dots,and contrast with the existed paper to prove the effectiveness and accuracy of the proposed method.Two methods are used to illustrate the effectiveness and accuracy of the proposed algorithm.First of all,the first method is to select the existing method of identifying umbral dots,and compare the recognition results with the results of proposed method to demonstrate the effectiveness of the algorithm.Secondly,another method is to statistically analyze some characteristics of umbral dots,including the equivalent diameters,lifetimes,intensities,size,trajectories and horizontal velocities,and to investigate the relationships between them.Finally,we contrast the results with previous results to prove the accuracy of the proposed method.It can be found that the proposed algorithm based on phase congruency technique can extract and identify umbral dots and their statistical characteristics accurately,and the statistical characteristics are well coincided with the existing results.It further proves the accuracy and effectiveness of the proposed method,and provides support and guarantee for further studying sunspots.