Intelligent network systems and security surveillance are used for smart city development. Manual and heritage technology is used in surveillance systems to detect speeds higher than usual in developing countries. This process can be automated with ease by leveraging the potential of image processing. Moving object detection is one of the essential tasks in image processing because of its prominent role in many real-world applications. Vehicle speed detection can be achieved by employing image and video processing methods. In this work, a unique approach is proposed for moving object detection and speed estimation based on an integrated approach in video sequence frames as opposed to the most commonly used RADAR and LIDAR devices for traffic law enforcement. Video data is collected and analyzed for speed in real-time without any sensor calibrations,thereby it removes any external hardware dependency requirements. Moving vehicles are segmented out by using frame-subtraction and masking techniques. The proposed algorithm tracks the time taken by the car to cover a predetermined fixed distance in order to calculate its speed.