This project proposes to use deep learning to enhance traffic safety as well as management through real-time detection of helmets, vehicle license plates, and rider distractions. Written in Python with the construction of the YOLOv8 algorithm, It detects if the rider wears a helmet classifies whether he/she is safe or unsafe and also checks whether a fine will be served for such a violation. Moreover, it detects distractions that the rider is involved in, like using a mobile phone while riding, serving enforcement for the traffic rule. The fine for each violation is displayed as text over the video so the commission of the offense can be observed. The system UI, which is done in HTML, CSS, and JavaScript, is created on the Flask web framework to produce things as smoothly as it can over user experience. It has three modes: image mode for photographs, video mode for early recorded videos, and Web camera mode for live detection. This means that it can be used for many purposes from image surveillance to real-time monitoring of the traffic. This project serves as a real tool for traffic authorities by spontaneously detecting helmet usage, deciding on the safety of the helmet, detecting rider distraction, identifying vehicle plates, and calculating fines for negligence. It gives a solution with high accuracy and flexibility that can be used for real-world cases contributing in the direction of better administration of traffic rules, decrease in rider complications, and better road safety.