Vehicle Detection Using CNN and YOLOv3

This research paper aims to develop a vehicle detection system using a YOLOv3 deep learning frameworks. The main objective of the study is to detect and localize the vehicles from the images or a video streams. The proposed algorithm is powerful enough to do the above task using the CNN(Convolutional Neural Networks) along with the YOLOv3(You Only Look Once) architecture to acheive the maximum possible outcome showcasing the results. This methodology involves us to train the YOLOv3 model on custom datasets divided into different types containing different environmental conditions. And another approach is to apply CNN based algorithm to greatly acquire the features accurately from the input images and also differentiating between the size, shape, wheels and orientation. Through extensive experimentation and evaluation on benchmark datasets, the effectiveness of the developed vehicle detection system is demonstrated. The results showcase a high detection accuracy and real-time performance, making it suitable for applications such as traffic monitoring, autonomous driving, and surveillance. The results of this research contribute to the field of computer vision and object detection by providing an accurate and efficient solution for the vehicle detection.