Gabor and Weber Local Descriptors performance in multispectral Earth Observation image data analysis

Abstract

During the last decades, both the satellite sensors and remote sensing imagery are evolving so fast that the methods and techniques of processing and analyzing Earth Observation (EO) data usually are staying one step behind. In the current paper, the authors goal was to prove the usability of Gabor and Weber Local Descriptors (WLD) in multispectral image classification. The extracted Gabor and WLD features were tested with Support Vector Machines (SVM), k-Nearest Neighbors (k-NN) and k-Means classifiers.