Combining multiple partitions into single ensemble clustering solution is a prominent way to improve accuracy and stability of clustering solutions. One of the major problems in constructing clustering ensembles is high computational complexity of the common methods. In this paper two computationally efficient methods of constructing ensembles of nonparametric clustering algorithms are introduced. They are based on the use of co-association matrix and subclusters. The results of experiments on synthetic and real datasets confirm their effectiveness and show the stability of the obtained solutions. The performance of the proposed methods allows to process large images including multispectral satellite data.

No abstract

No abstract

An effective method of training set extension for aerospace images classification is proposed. The method is based on mean shift procedure with respect to spatial information. It allows considering the unlabeled data structure. The results of experimental study using the Salinas hyperspectral image are presented, proving the effectiveness of the proposed method.