The ability to recognize and track cars is a vital part of traffic surveillance systems, and is critical for the efficient management of traffic and the safety of drivers and passengers. Finding and following the path of vehicles is the primary goal of this research. The goal of this research is to develop a system that can recognize cars in photos and videos automatically. Deep Learning, a technique that may include fuzzy logic, neural networks, and evolutionary algorithms, has various applications, one of which is the detection and tracking of automobiles. We will utilize YOLOv8 and the Kalman filter or the DeepSORT algorithm to follow vehicles throughout the course of a movie using deep learning for this project. This is the main topic of the article. Afterward, the DeepSORT algorithm forms the backbone of the vehicle detecting system. By adding focus loss as an optimization component to the original DeepSORT method, we are able to improve the performance of feature extraction. Therefore, the procedure begins with a series of training procedures applied to photos from the publicly accessible road vehicle dataset. After that, we employ YOLOv8 and the DeepSORT algorithm to follow the vehicle identification model, and their combined efforts show how effective they are. To locate it, one must examine the detection rates achieved by both models on different types of automobiles. The fundamental objective of this study is to develop an automated technique for detecting and tracking autos in both static and dynamic scenes. Once the network model has been trained, it is applied to the analysis of the camera video from the car, and the detection performance is evaluated experimentally. The study's findings show that the technique's success rate in recognizing automobiles has grown to 98.48%. In addition, using DeepSORT as the vehicle tracker results in decreased mistake rates.