Vehicle speed estimation is vital for intelligent transportation systems, and the use of digital image processing techniques like YOLOv8 and OpenCV has garnered significant attention. This paper introduces a novel approach that combines OpenCV and YOLOv8 to estimate car speeds within a designated Region of Interest (ROI). A dataset comprising a video from a traffic surveillance camera is collected and utilized to train and evaluate the car speed estimation system. OpenCV is employed to process the video data, while the YOLOv8 model is employed in the detection of objects to identify cars within the ROI. The ‘cv2.EVENT_MOUSEMOVE’ constant from the OpenCV library is employed to obtain the ROI. By analyzing the car’s motion within the ROI, the vehicle’s speed is accurately estimated. The proposed approach offers advantages over traditional methods. It provides a cost-effective solution with improved coverage compared to radar and laser-based systems. Accurate car speed estimation within the specified ROI is crucial for effective traffic management and the development of intelligent transportation systems. Evaluation results show that the suggested strategy is effective. The system successfully estimates car speeds within the defined ROI, contributing to accurate speed limit detection and enhancing traffic management efficiency.