Improving the operation of the detection and classification of vehicles on satellite images by reducing the search area using geo-information about roads

A method is proposed for accelerating and increasing the accuracy of the algorithm for detecting and classifying vehicles on satellite images. The method is based on the reduction of the search area by extracting and processing geographic information about the location of roads on the ground. Using this method significantly reduces the amount of data processed by the detection and classification algorithms, and as a result, the execution time and error of these algorithms are reduced.