Штучний інтелект

Анотація: The thermal imager, which helps operator to orient on area, detect and recognize target in the night conditions, is included as rule in structure of battle modules of unmanned complexes. Partially these tasks are placing on electronics at implementation of artificial intelligence into unmanned complex. But classical thermal imager optical system forms on digital photodetector the two-dimensional image of surrounding space. It hinders the target recognition by battle unmanned complex artificial intelligence at night conditions. It is well known from fundamental theory of phenomena of light reflection and refraction on border “medium-air”, which is described by Fresnel’s low for metals and dielectrics, the polarization rate of thermal objects depends on position of normal of object elementary small surface relatively to direction of its observation (relatively to optical axis of observation device). At the same time, polarization rate of surface own irradiation is increasing at rise of angle between irradiation direction and normal to irradiation surface. And that angle is changing within wide diapason, for example, during exploration from unmanned complex board. On that base there is possibility to determine the third coordinate of target elementary small surface and to image a 3D target on thermal imager display screen or to recognize a target by unmanned complex artificial intelligence. Authors show, including of infrared polarizer, rotating around the thermal imager optical system axis on fixed angles, into thermal imager structure permits to determine the third target coordinate and to increase a probability for its recognition by unmanned complex artificial intelligence in that case.

Ключові слова: unmanned complex, unmanned complex, thermal imager, infrared polarizer, recognition probability, heat target

Посилання:

1. Dovgopolyy A., Biloborodov O., Senatorov V. (2021). Shtuchnyy intelekt bojovogo nazemnogo robotyzovanogo kompleksu. Tezy dopovidej ХХІ mizhnarodnoi naukovo-tehnichnoi konferentsii «Shtuchnyy intelekt ta intelektualni systemy (AIIS 2021)», Kyiv, Instytut problem shtuchnogo intelektu.
2. Kolobrodov V.G., Lykholit M.I. (2007). Proektuvannia teplovizijnyh i televizijnyh system sposterezhennia. Kyiv: NTUU «KPI».
3. A.s. SU 1667273 А1. (1991). Sposob teplovizionnogo raspoznavanija formy objektov. Bul. 28.
4. Pat. RF 2024212 С1. (1994). Sposob raspoznavanija formy objektov. Bul. 22.
5. Pat. RF 2099759 С1. (1997). Teplovizionnyj sposob raspoznavanija formy objekta i ustrojstvo dlia ego osushchestvlenia. Bul. 35.
6. Tymkul O.V., Tymkul V.M., Ushakov O.K. (1999). Metodika opredelenia objomnoj formy objektov na osnove polirizatsionnoj kombilirivannoj termogrammy. Opticheskij zhurnal. 2, 54-59.
7. Fesko Yu.A. (2014). Razrabotka i issledovanie optiko-elektronnyh metodov opredelenija trehmernoj formy objektov. Avtoref. dissertatsii na soiskanije uchenoj stepeni k.t.n. Novosibirsk: SGGA.
8. Frederik S. Leira, Håkon Hagen Helgesen, Tor Arne Johansen, Thor I. Fossen. (2021). Object detection, recognition, and tracking from UAVs using a thermal camera. Journal of field robotics. 2, 242-267. https: //doi.org /10.1002/rob.219.
9. JM. Lloyd. (1978). Sistemy teplovidenija. М.: «Mir».
10. М.Born, E.Wolf. (1973). Osnovy optiki. 2-e izd., per. s angl. pod red. G.P.Motulevich. M.: Nauka
11. Ivanov V.P., Kurt I.V., Ovsjannikov V.А. i dr. (2006). Modelirovanije i otsenka sovremennyh teplovizionnyh priborov. Kazan’: Otechestvo.
12. Mikitenko V.S., Plavi V.E.S. (2016). Pokazatel kachestva izobrazhenij v mnogokanalnyh optiko-elektronnyh sistemah s poljarizatsionnym kanalom. Mezhdunarodnyy nauchnyy zhurnal, 3, 78-83. Available: https://ela.kpi.ua/handle/123456789/37779.
13. Fesko Yu.A., Tymkul V.М., Tymkul L.V. i dr. (2011). Мatematicheskaja model opredelenija trjohmernoj formy objektov na osnove dvuh poljarizatsionnyh termogramm. Sb. materialov VII Mezhdunarodnogo kongressa «GEO-Sibir-2011». Novosibirsk: SGGA.
14. Pat. RF 2431936 С1 (2011). Sposob teplovizionnogo raspoznavania formy objektov. Bul. 29.