Real-Time Traffic Detection using Horizontal and Vertical Scanning

Abstract

Traffic congestion on city road networks is one of the main issues to be addressed by today's traffic management schemes. Traffic congestion at times leads to delay in emergency services (i.e. Ambulance, Firefighter, Police, etc.) and most of the time causes inconvenience to commuters. In this paper a new traffic detection method is proposed which is based on the horizontal and vertical scanning of video frames to obtain accurate vehicle detection. Traffic congestion is measured in terms of traffic intensity which helps to ascertain whether the traffic at a given point is low, medium or high. The proposed method shows better result as compare to other methods in terms of accuracy.Keywords: Intelligent Traffic management, Emergency Vehicle, Morphological Operations, Horizontal and vertical scanning and video processing. I. INTRODUCTIONWith increase in population and corresponding increase in vehicular traffic leads to road congestion. Such traffic congestion leads to many major problems and challenges in the urban areas and populated cities. This causes wastage of time leading to missing opportunities, frustration and delay in reaching destination.Traffic loads depend heavily on parameters such as time, day, season, weather conditions and unpredictable situations such as accidents, special events or construction activities. If these parameters are not taken into account, the traffic control system will create delays. To solve the problem of congestion, new roads are built. The only disadvantage of making new roads on the facilities is that it makes the environment more congested. So, for that reason, it is necessary to change the system rather than make new infrastructure twice. A traffic control system can solve these problems by continuously detecting and adjusting the timing of traffic signals according to the actual traffic load such a system called as intelligent traffic control system. The advantages of building an intelligent traffic control system is to reduce congestion; operating costs; provide alternative routes for travelers, increase the capacity of the infrastructure.Most urban traffic is controlled by sensors and cameras which needs to be installed on major roads and streets. The existence of an automatic traffic detection system will assist in determining the traffic [1]. These systems extract information from the larger traffic issue and help us decide to improve traffic policy [3,4]. The goal of this current research is to develop an automatic vehicle counting system, which can process videos recorded from stationary cameras over roads e.g. CCTV cameras installed near traffic intersections / junctions and counting the number of vehicles passing a spot in a particular time for further collection of vehicle / traffic data. The paper aims to automate the traffic control system on highways and streets to determine Traffic Density , identify streets, roads in order to count the cars and monitoring of roads. Figure1 shows a sample image of urban traffic.

Improvement of Alert System against Tampering and Theft in Surveillance Cameras

“…As mentioned before, the dissimilarity takes place between two frames. [6] But first; the frames should be pre-processed to make the comparison more accurate, and this performed by using erosion process. The basic effect is to erode away the boundaries of regions of foreground pixels.…”

“…This approach is based on comparing two sequenced images (frames) in live video stream provided by camera, which will detect any suspicious movements regarding to selected detection threshold, but in case of changing the direction of the camera, the situation is different, here the comparing of the two sequenced frames is not efficient, so instead, the comparison is performed between reference frame k (which will be changed simultaneously) and frame (k+15), this because sometimes the thief try to move the direction of the camera slowly to prevent the system from detecting them, but the approach presented in this paper will set another threshold to accomplish such tampering. Equation (1) shows how to compare two frames after preparation [6]. 𝐷 (𝑥,𝑦) = 𝐼 𝑡 (𝑥, 𝑦) − 𝐼 𝑡−𝑛 (𝑥, 𝑦) > DT ….. (1) Where 𝐷 (𝑥,𝑦) represents the difference between two frames in pixel(𝑥, 𝑦), 𝐼 𝑡 (𝑥, 𝑦) is the intensity of pixel (𝑥, 𝑦) in grayscale, 𝐼 𝑡−𝑛 (𝑥, 𝑦) is the intensity of pixel (𝑥, 𝑦) according to n in grayscale, DT is the detection threshold and n value defined as in Equation 2.…”