The Advancement Of An Automatic Solar Eruptive Event Detection System (seeds) To A Near Real-time System
Abstract
We have been working on a tool to automatically detect, track and catalogue Coronal Mass Ejections as observed by the LASCO C2 instrument on board the SOHO spacecraft. The software developed follows a simple algorithm, which is based on decomposing 2-D Coronagraph images to 1-D and looking for outstanding bright regions. Thresholds determine these regions, and in time sequences, CMEs can be located and tracked. Many improvements to the algorithm have been made in recent months dealing with the preprocessing of the input images as well as the tracking of CME detections. Most importantly the system has advanced to the point that it can make detections of CMEs in near real-time. What is meant by near real-time detection is; first, the downloading images, as they are made available by the instrument team, and second, the processing and reporting of CME events as they occur on a daily basis. The true positive rate of detections is over 75% as compared with the CDAW CME catalog, and will continue to improve with further experimentation of threshold values. Output from the SEEDS algorithm can be found at http://solar.gmu.edu/research/autocme/ (Still showing preliminary results). Currently, the software is being adapted to accept STEREO images as input, preliminary test with STEREO COR2 data show promising results.
- Publication:
-
American Astronomical Society Meeting Abstracts #210
- Pub Date:
- July 2007
- Bibcode:
- 2007AAS...210.2929O