Identification of Sunspots on Full-Disk Solar Images Using Wavelet Analysis
A new method to detect sunspots on full-disk solar images, recorded in an optical wide spectral band and in the line of Ca ii K1, is presented. It is based on the compact wavelet transform and a process that automatically identifies the noise in the images of the Sun. This method estimates also its center and the limb positions with a precision of a few milliarcseconds. Precisions of 1 and 8 milliarcseconds are, respectively, found for the disk center position and the radius of simulated solar images. The results obtained on sunspot identification on solar images recorded during April 2002 at Meudon Observatory (France) reveal that we are able to detect 3.35 % additional sunspots compared to a manual method. Our method preserves the sunspot areas better than 95 %.
KeywordsEdge detection Gaussian filters Sunspots Sun radius Wavelet analysis
This work has been developed with the support of Centre National d’Etudes Spatiales (France), CNRS, and the French Foreign Affair Ministry under the framework of scientific cooperation contract TASSILI between France and Algeria.
- Győri, L., Baranyi, T., Turmon, M., Pap, J.M.: 2002, In: Wilson, A. (ed.) Proc. SOHO 11 Symp.: From Solar Min to Max: Half a Solar Cycle with SOHO, ESA SP-508, 203. Google Scholar
- Meftah, M., Meissonnier, M., Irbah, A., Abbaki, S., Assus, P., Bertran, E., et al.: 2010, In: Oschmann, J.M. Jr., Clampin, M.C., MacEwen, H.A. (eds.) Space Telescopes and Instrumentation 2010: Optical, Infrared, and Millimeter Wave, Proc. SPIE 7731, 773145. Google Scholar
- Zharkov, S., Zharkova, V., Ipson, S., Benkhalil, A.: 2005, EURASIP J. Appl. Signal Process. 15, 2573. Google Scholar