Алгоритм морфологической обработки радиолокационных изображений и автоматического обнаружения объектов по радиолокационной тени

Abstract

The article substantiates the sequence of operations for processing radar images, which provide automatic detection and selection of objects by their radar shadow. Unlike existing detection algorithms with a constant level of false alarm (CFAR-Constant False Alarm Rate), it is suggested to apply morphological operations on an inverted binary image. The parameters of morphological operations that ensure the elimination of distortions of the radar shadow of the object and the exclusion from the “area of interest” (ROI-Regions of Interest) of objects whose radar shade differs from the reference ones (calculated based on the size of the object and the conditions for its recording) are determined. The algorithm works in the complete absence of information about the parameters of the probability distribution density of the sample counts of the object and the background. The results of processing radar images of real objects located on a natural background are given