Automatic aurora images classification algorithm based on separated texture
Rong Fu, Jie Li, Xinbo Gao, Yongjun Jian · 2009
In order to resolve the problem incurred by low efficient manual classification of tremendous aurora images, an automatic aurora images classification system for huge dataset application is proposed. First, static aurora images are decomposed into texture part and cartoon part with a method called Morphological Component Analysis (MCA). Then features extracted from texture part are classified by three classification methods: nearest neighbor (NN), Support Vector Machine (SVM) with RBF kernel and SVM with linear kernel. The experiment exhibited the classification accuracy improved by 10%, of which, the SVM with linear kernel is much faster and is therefore suitable for massive data processing.