Methods for Auroral Oval Segmentation in Images (Part 1)

The paper provides an overview of studies conducted between 2007 and 2014 on auroral oval segmentation in images. Full sky maps and ultraviolet images were used as input data. The following methods were considered: a modified version of the contour-label isolation map algorithm, the linear least squares randomized Hough transform (LLS-RHT) method, methods based on pulse-coupled neural network (PCNN), k-means histograms, adaptive minimum thresholding errors (AMET), fuzzy local information c-means (FLICM) clustering algorithm, level setting method with shape initialization and intensity adaptation.