Khlamov S. Computational methods of data processing for detection of objects with near-zero apparent motion

The object of research is the objects with near-zero apparent motion in the series of CCD-frames. The purpose of research is to improve the quality indicators of detection of the objects with near-zero apparent motion due to the development of new and refinement of well-known computational methods of data processing. The methods of research are methods of the mathematical statistics theory, theory of the optimization, dispersion analysis, statistical theory of hypothesis testing and estimation of parameters that allowed to develop computational methods for detection of the objects with near-zero apparent motion; theory of statistical modeling and processing of natural data, which allowed to develop methods for analysis of the quality indicators of detection of the objects with near-zero apparent motion. Theoretical and practical results consist of the following. The computational methods for data processing, developed in the dissertation, allow increasing the probability of detection objects with near-zero apparent motion in the series of CCD-frames versus to traditional methods. Scientific novelty of the results lies in the fact that first proposed a computational method for detection of the objects with near-zero apparent motion that based on the criteria of velocity significant with decision rules of statistical testing of the velocity significant of object using Student t-criteria and Fisher f-criteria. Its using maximizes the conditional probability of a true detection of the objects with near-zero apparent motion in constantly changing observation conditions. Improved the substitutional computational method for detection of the objects with near-zero apparent motion that based on the substitutional decision rules of the detection of apparent motion, using of which is justified by the absence of a priori data on the parameters of the object’s apparent motion in case of using the different values of the dispersion of the position measurements in the series. Improved the method for analysis of the quality indicators of detection of the objects with near-zero apparent motion that based on statistical and natural modeling, using of which allows to determine quality indicators of detection of the objects with near-zero apparent motion by developed computational method for detection in case of using the real distribution law of the position measurements errors of object. The thesis work was carried out in accordance with the plan of research work of Kharkov National University of Radio Electronics in the framework of research work №307 «Dynamic intellectual analysis of sequences of fuzzy information under conditions of significant uncertainty based on hybrid systems of computational intelligence» (№SR 0116U002539, act of 10.10.2016). The developed computational methods of data processing implemented in the educational process in the Kharkov National University of Radio Electronics (act of 20.09.2016), at the observatory Odessa-Mayaki (code in the Minor Planet Center - 583) (act of 17.10.2016) and in the laboratory of space research of Uzhgorod National University (act of 27.09.2016). The developed computational methods can be used in data processing software for detection of the objects with near-zero apparent motion, for example in the programs of the automated detection of small bodies of the Solar system (asteroids and comets) in the series of CCD-frames.