Diffuse Objects Extraction in Coronal Holes Using Active Contour Means Model
Abstract
This paper presents the application of active contour models (Snakes) for the segmentation of diffuse objects from coronal holes on the sun imaging telescope by means of active contour. In this paper the partition is less important and the focus is on the extraction of diffuse objects, namely coronal holes in EIT (Extreme Ultraviolet Imaging Telescope) images. Coronal holes are regions of low-density plasma on the sun that have magnetic fields that open freely into interplanetary space. The proposed method is based on level set evolution. Since an active contour is a dynamic spline that can be adjusted to fit onto the boundary of the object based on energy minimization, and this method independent of selective region which we want to extract, hence using this method improved the result of segmentation. In addition to extracting diffuse objects of coronal holes, a new work that has been done to improve results, is that speed algorithm by implementing two methods: Reinitialization and Narrow band which related to the level set method has been used. Received images from satellites have large size so this method is very effective on these images.
Keywords
Active contour model; Coronal holes; Parametric model; Level set; Reinitialization; Narrow band