2518-1092Research result. Information technologies2518-109210.18413/2518-1092-2024-9-4-0-53667ARTIFICIAL INTELLIGENCE AND DECISION MAKING<strong>FORMATION OF AN ENSEMBLE OF MODELS FOR RECOGNIZING OBJECTS IN RADIOGRAPHIC IMAGES&nbsp;BASED ON INFORMATION THEORY</strong><strong>FORMATION OF AN ENSEMBLE OF MODELS FOR RECOGNIZING OBJECTS IN RADIOGRAPHIC IMAGES&nbsp;BASED ON INFORMATION THEORY</strong>BasovOleg OlegovichBasovOleg Olegovicho.basov@acti.ruTrukhachevAndrey AleksandrovichTrukhachevAndrey AleksandrovichSobolevYuri IgorevichSobolevYuri IgorevichSeleznevVladimir VladimirovichSeleznevVladimir Vladimirovich20249400

The paper considers the main functional approaches to the construction of image processing schemes for various spectral ranges, energies and scales. Their effectiveness is estimated from the point of view of the information value determined by the degree of achieving the goal of the non-destructive testing system - detection of all possible objects of interest with a minimum number of images. It is established that in the absence of the possibility of optical (visual) control and the technical possibility of obtaining radiographic images at different emitter energies, it is possible to improve the quality of object recognition by using an ensemble of recognition models. This conclusion is confirmed by a specific example demonstrating an improvement in the quality of recognition of objects of interest by a two-level ensemble of Yolo8 models by 12-18% compared to their recognition using a single neural network model.

The paper considers the main functional approaches to the construction of image processing schemes for various spectral ranges, energies and scales. Their effectiveness is estimated from the point of view of the information value determined by the degree of achieving the goal of the non-destructive testing system - detection of all possible objects of interest with a minimum number of images. It is established that in the absence of the possibility of optical (visual) control and the technical possibility of obtaining radiographic images at different emitter energies, it is possible to improve the quality of object recognition by using an ensemble of recognition models. This conclusion is confirmed by a specific example demonstrating an improvement in the quality of recognition of objects of interest by a two-level ensemble of Yolo8 models by 12-18% compared to their recognition using a single neural network model.

cascade of modelsobject recognitioninformation theoryvalue of informationnon-destructive testingcascade of modelsobject recognitioninformation theoryvalue of informationnon-destructive testingСписок литературыGOST 18353-79. Nondestructive control. Classification of types and methods. Date of introduction 01.07.1980. (in Russian).Afonin P.N. Working with inspection X-ray machines. SPb.: IC Intermediya, 2017. &ndash; 240 p.Kogan I.M. 1981. The applied information theory. Moscow, Radio and communication, 216 p. (in Russian).Osipenko A.A., Basov O.O. Modeling of non-destructive testing systems of electronic modules based on information theory // Belgorod State University. Scientific Bulletin. Series: Economics. Information technologies. &ndash; 2018. &ndash; T.45. &ndash; №1. &ndash; P. 93-102. DOI: 10.18413/2411-3808-2018-45-1-93-102. (in Russian).Grigorov M.S., Basov O.O. 2015. Method of formation of the x-ray multiimage of a product of microelectronics with heterogeneous structure // Belgorod State University. Scientific Bulletin. Series: Economics. Information technologies. &ndash; 2015. &ndash; № 7&nbsp;(204). &ndash; Issue 34/1. &ndash; P. 67-72. (in Russian).Osipenko A.A., Ignatenkova O.A., Grigorov M.S., Basov O.O. 2017. Justification for need of combined use of automatic optical inspection and non-destructive x-ray control of electronic modules // Research result. Information technologies. &ndash; Т. 2. &ndash; № 2. &ndash; P. 3&ndash;8. (in Russian).