Information-extremе machine learning for object identification on the terrain

Journal Title: Журнал інженерних наук - Year 2016, Vol 3, Issue 1

Abstract

The article deals with the usage of a SURF local descriptor of key fragments to create a global descriptor BoF for objects of interest on terrain within task of recognition of armored technique in the controlled territory using images of air reconnaissance. The method of optimization of the input mathematical description of the information-extreme classifier trained on dataset, which consist of global descriptors BoF, is proposed. The optimal in information understanding dimensionality of a global descriptor is determined by iterative procedure, which includes k-means clustering of key fragment's SURF-vectors, training dataset creation and information-extreme machine learning of the classifier. The offered algorithm of information-extreme machine learning implements the adaptive coding of values of primary features using multilevel system of control permits, and creation of hyperspherical containers of classes in binary space of secondary features with sequential optimization procedures. It was suggested to use the rated modification of S. Kulbak's information measure, which is a function of false omission rate and positive predictive value of decision-making and it also allows machine learning on imbalanced dataset.

Authors and Affiliations

V. Moskalenko, A. Korobov

Keywords

Related Articles

Clarification of the Recent Scientific Approaches in Magnetic Water Treatment

The paper focused on study of influence magnetic water treatment (MWT) mechanism on physicochemical properties of aqueous solutions taking into account nature of changes in kinetics of chemical reactions. The theoretical...

Dynamic Simulation of Heat Transfer through External Building Constructions

In the paper, some approaches to dynamic modeling of heat transfer through external building constructions are presented. The model of heat transfer dynamics through a wall was designed on the base of mathematical descri...

The hypothesis of formation of the structure of surfaced metal at the surfacing based on the application of the prognostic algorithm of control the electrode wire speed

The growth of a drop in the process of surfacing by a consumable electrode is characterized by a linear dependence of the current change on time. A hypothesis has been put forward, according to which a reduction in the f...

Systematic Presentation of Ritz Variational Method for the Flexural Analysis of Simply Supported Rectangular Kirchhoff–Love Plates

In this work, the Ritz variational method for solving the flexural problem of Kirchhoff–Love plates under transverse distributed load has been presented systematically in matrix form. An illustrative application of the m...

Characterization and Degradation of Viton Fuel Hose Exposed to Blended Diesel and Waste Cooking Oil Biodiesel

Degradation and material incompatibility between biodiesel and fuel system are the major concern associated with the adoption of biodiesel. In this research, effects of different mixture of waste cooking oil biodiesel/di...

Download PDF file
  • EP ID EP241741
  • DOI -
  • Views 71
  • Downloads 0

How To Cite

V. Moskalenko, A. Korobov (2016). Information-extremе machine learning for object identification on the terrain. Журнал інженерних наук, 3(1), 1-7. https://europub.co.uk/articles/-A-241741