The proposed algorithms that have been used in this Project are Saliency, Background Subtraction, K-Means, Dilation, Erosion, Blob, and YOLO. The YOLO v3's accuracy in the project is faster and more accurate, based on the coco dataset. It contains 53 convolutional layers that are efficient in recognizing the objects in the video. The dilation applied in the Vehicle Detection and Classification system here to enhance detection accuracy in using the YOLO v3. The object's score is detected between the range of 0 to 1, which objects detected with the score closer to 1 is more accurate depending on the probability.