A.V. Кorennoy, S.A. Kuleshov

In this paper we propose a new way of describing the portraits of radar images objects obtained in remote sensing of the earth's surface in the form of mathematical models. The resulting models are needed for automated processing radar images. For the automated analysis system radar images were necessary to have a bank of reference images of the objects interest. Currently in use as reference pictures radar portraits of objects. Using radar portraits of objects as references has its drawbacks: the need to store large amounts of data, significant computational cost of processing the complexity of their receipt. It is therefore proposed to use the model of radar images objects. As the real radar images of objects do not have clear boundaries, i.e. change the brightness of the object described by a monotonic function as a mathematical model of the shape objects is proposed to use a two-dimensional decaying exponential, an indicator of which are the lines of higher orders. The proposed approach allows us to create a database of images objects of reference models that accurately describe the actual objects.

 

1. Zuev Ju.S., Reshetneva T.G. Primenenie metodov distanczionnogo zondirovaniya v geoinformatike (na primere snimkov kamery' Aster) // Geoinformaczionny'e sistemy'. 2003. № 1(5). S. 57-65.
2. Radiolokaczionny'e sistemy' vozdushnoj razvedki, deshifrirovanie radiolokaczionny'x izobrazhenij: uchebnik / pod red. L.A. Shkol'nogo. M.: VVIA im. prof. N.E. Zhukovsko­go. 2008. 532 s.
3. Duda R., Xart P. Raspoznavanie obrazov i analiz sczen: per. s angl. G.G. Vajnshtejna. M.: Mir. 1976. 512 s.
4. Savelov A.A. Ploskie krivy'e. Sistematika, svojstva, pri­meneniya. M.: Gosudarstvennoe izdatel'stvo fiziko-matematicheskoj literatury'. 1960. 294 s.
5. Obnaruzhenie, raspoznavanie i opredelenie parametrov obrazov ob''ektov. Metody' i algoritmy' / pod red. A.V. Korennogo. M.: Radiotexnika. 2012. 112 s.