Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back Two-dimensional objects recognition and determination of their parameters on radar images of the earth surface A.V. КORENNOY S.A. KULESHOV An important task of monitoring the earth’s surface is the detection and recognition of interest objects in real-time. To reduce the time data interpretation invited to perform preprocessing radiogologrammy aboard the carrier, ie determine if a received signal image search objects. The proposed approach to solving the problem of joint recognition and parameter estimation of images of objects requires a priori information in the form of objects models to be identified. As a mathematical model of image objects is proposed to use two-dimensional decreasing exponent, which are an indicator of higher-order line. The radar picture images of objects may be positioned at different angles and their location will always coincide with the reference image to be used as an a priori model. Therefore, even if the same size and shape of the reference image and a real object, but varying their orientation with respect to each other, the desired object recognition may not be possible. On the basis of the Bayesian approach is designed bifunctional nonlinear algorithm joint recognition objects and estimation of their parameters in the radar image the Earth’s surface. The mathematical model of radar images of objects. The influence of the relative orientation of the object and the reference image on the result objects recognition and estimation of its parameters considered. Keywords: radar image, models of images, parameters estimate of objects, pattern recognition, Bayesian approach. Contacts: E-mail: korennoj@mail.ru Pp. 57-64.