Automatic Tracking of Coronal Mass Ejection using STEREO Red-colored RGB Coronagraph Images

摘要

In this paper we propose a new technique to segment Coronal Mass Ejection (CME), track its leading edge and estimate the propagation parameters using STEREO A/B SECCHI COR2 red-colored RGB images. The algorithm consists of two modules: pre-processing and classification. The pre-processing module uses a red component of the images to segment CME and produces a set of running-difference binary images. This set is then fed into the classification module that transforms images into polar coordinates, detects CME front edge, and computes the CME propagation parameters. The method was validated using total 30 CMEs, 15 STEREO A and 15 STEREO B events captured in the period from 1 May 2008 to 31 August 2009. The results demonstrate that the proposed method is effective for CME tracking and estimation of propagation parameters. The proposed method can be integrated with the triangulation technique to estimate propagation of CME in three-dimensional space.