Deep Learning Based Vehicle Detection with Images Taken from Unmanned Air Vehicle
Mustafa Böyük, Ramazan Duvar, Oğuzhan Urhan
Abstract
Especially in recent years, unmanned aerial vehicles (UAV) have been used in many platforms for military, commercial and entertainment purposes. Thanks to high computational graphics cards, many computer vision applications based on deep learning can work in real time on unmanned aerial vehicles. In this study, fast and high-performance object detection algorithms Faster R-CNN, YOLOv3-Tiny and SSD algorithms are used to detect vehicles in image taken from UAVs. Thus, the positions of vehicles driving in traffic are found automatically.
Topics & Concepts
Computer scienceObject detectionDeep learningArtificial intelligenceComputer visionDroneGraphicsComputer graphicsReal-time computingComputer graphics (images)Pattern recognition (psychology)BiologyGeneticsAdvanced Neural Network ApplicationsVideo Surveillance and Tracking MethodsVehicle License Plate Recognition