Commutation is an important aspect of our everyday lives. Vision-based vehicle detection and classification has grown in prominence in the field of intelligent transportation systems research. Unmanned aerial vehicles (UAVs) for civilian remote sensing have sparked a lot of interest in recent years because they benefit the community in a variety of ways, such as recovering missing vehicles, avoiding traffic areas, and tracking vehicle quantity on road to plan the roads and establish new rules. In challenging traffic situations, vehicle detection is often used. Researchers have devised many approaches to these problems. Some of them produce decent results but are computationally costly and fail in some conditions. This detailed literature review delves into the specifics of the most accurate and efficient methods for recognizing and identifying cars. This paper focuses on many common techniques for vehicle detection. A few of the algorithms compared in the study include Convolutional Neural Network, You Only Look Once, Support Vector Machine, Faster R-CNN and Single Shot Detector. We can determine whether the approach provides improved precision for vehicle detection based on the comparison. From the contrast, we can analyse which method gives enhanced precision for vehicle detection.