The study uses a variety of deep learning methods to address the issue of vehicle detection using aerial pictures. The photographs are taken by unmanned aerial vehicles (UAVs), providing us with a clear picture of every vehicle. The primary goal of this endeavor is to identify every kind of vehicle that is visible in the picture, making it helpful for a variety of tasks like counting cars, parking system empty slots, and analysis of traffic. Numerous deep learning methods are in use to detect automobiles. Deep learning techniques like You Look But Once (YOLO), Faster region-based convolutional neural network (Faster R-CNN), convolutional neural network–support vector machine (CNN–SVM), and single shot detector (SSD) are among them. Comparing the strategies employed by CNN–SVM, YOLO, SSD, and Faster R-CNN algorithms is the research's main goal. This report highlights both differences and parallels across various approaches and discusses the difficulties that lie ahead.