Detection of small-scale features in the solar photosphere using segmentation algorithms

CABELLO, I.; BALMACEDA, L. A. ;; DOMINGO, V.; BLANCO-RODRÍGUEZ, J.; BONET, J.A.

Workshop; Solar Information Processing WORKSHOP VI; 2012

The study of small magnetic structures in the solar photosphere is of great relevance for the understanding of the global behavior of the Sun. Because of the small spatial and temporal scales involved, the use of high resolution images is fundamental for their study. In order to obtain such images, sophisticated computational techniques that compensate for the atmospheric degradation and telescope aberration have been developed, improving in this way the spatial resolution. In this work, we use G-band images obtained with the 1m-Swedish Solar Telescope located at La Palma (Canary Islands, Spain) and restored with MFBD (Multi-Frame Blind Deconvolution), a technique that combines multiple images acquired in a short time interval. The resulting images have a resolution close to the diffraction limit of the telescope (0.''1) allowing the study of the properties of very small bright structures present in the inter-granular lanes in the solar photosphere, known as Bright Points. In order to detect such small-scale features, algorithms based on intensity thresholding and image-segmentation are used. In this work, we present the results obtained after applying two algorithms of this kind: the one designed by Sanchez Almeida et al. (2004) and MLT4 developed by Bovelet and Wiehr (2007). We compare their performance based on the number of features successfully detected, the time necessary to do the task and the level of participation of the observer when defining the intensity thresholds and determining whether a given structure corresponds or not to a Bright Point.