Genetic Algorithm and DWT Based Multilevel Automatic Thresholding Approach for Vehicle Extraction

Vehicle Extraction from aerial images is an important research topic in surveillance, traffic monitoring and military applications. In this paper, an approach based on Automatic Multilevel Thresholding has been proposed for extracting vehicles from aerial imagery. The approach combines Genetic Algorithm with DWT to make segmentation faster and geometric feature of vehicles for vehicle extraction. This algorithm analyses the color and connected properties of pixels to extract the outline of vehicles. In this research, UAV colour imagery is examined experimentally. After analysis, it is examined that proposed method provides the vehicle position accurately.