Method for detection of vehicles in digital aerial and space remote sensed images

In this paper we propose a method for automatic detection of the location of automobiles in the aerial images and images of advanced space-based Earth remote sensing means (ERS). The proposed method uses various groups of features (geometric, textural, structural) for a description of objects depicted in the images and their preliminary analysis, as well as solving list as an algorithm of testing of hypothesis of the presence of an object. The final positioning of objects in the image is performed with the original consecutive procedure. The proposed method on real urban area photographs provides a high quality score of vehicles correct detection with a slight (a few) number of falsely detected objects.