Smart Computing and Systems Engineering - 2018 (SCSE 2018)
Permanent URI for this collectionhttp://repository.kln.ac.lk/handle/123456789/18937
Browse
3 results
Search Results
Item Vehicle type validation for highway entrances using convolutional neural networks(International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Juwanwadu, L.N.W.; Jayasiri, A.Vehicle type validation for Highway entrances using convolutional neural networks is an approach taken to automate the highway toll systems of Sri Lanka. Available automated highway toll systems in the world use sensor-based validation systems to validate the vehicles that are entering the highways. Mainteneance cost of these systems is high. A vision-based validation system has not been implemented, as yet. This paper introduces a vision-based method to validate vehicles for highway systems which can reduce the cost while increasing the efficiency and safety. A Convolutional Neural Network (CNN) model was developed to achieve this objective. The CNN model employed here uses a binary classification to categorize vehicles as allowed vehicles and non-allowed vehicles for entering the highway. The model developed here showed 86.69% accuracy. The model was manually tested for different vehicle types using a GUI based application and all the test images were successfully classified into their classes.Item Traffic through - An effective right-turn-bay of signalized intersections for busy hours(International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Seneviratne, W.Y.H.; Jayasiri, A.Sri Lanka, as a developing country, the traffic congestions are becoming severe day by day because of the population and the economic background. To overcome the congested areas, the traffic controlling light systems were introduced and used in most of the junctions. Existing traffic light system is based on the fixed cycle times for each phase, depending on the environmental conditions, geometry design of the junction and traffic movements of the particular junction. Among the most of the features at the road intersections, the feature called “Right-Turn-Bay” is an extra segment of lane which was introduced to the vehicles which are supposed to proceed right turning movement at the intersection. Because of the limited area of that lane segment, it can be filled easily. The proposed solution is to overcome the overflowing problem in the right-turn-bay by giving an extra cycle time period for the right turnings in busy time. For this real-time process, the image processing techniques were used for a video sequences, captured by a video camera to detect the arrival of vehicles at the bay and its count was gathered using the pixel-wise detection and blob tacking of the vehicle. The traffic light was controlled as usual and the additional functionality was to control the junction considering the maximum count and the current count at the bay. The controller is to decide whether the traffic should leave the bay or not by considering its parameters.Item Intelligent traffic controller using image processing(International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Fernando, J.J.R.S.; Jayasiri, A.Traffic congestion has become significant problem in recent years with the ever increasing number of vehicles and poor management of traffic. Traffic patterns are not constant throughout the day. They are changing from time to time. Since present traffic controllers have fixed time intervals for signal lights, they could not provide a better solution. Computer vision can be used to create an intelligent traffic controller which can adapt its time intervals according to the real traffic. Several studies have been carried out based on the concept of real time image processing to manage the traffic. In current traffic controllers, wastage of effective green time is occurred, as many times fixed green time period which is assigned for a phase is larger than it actually needs. Hence the other roads at the intersection have to wait in vain, with more traffic, until that fixed green time period is over. In the proposed method real time traffic image sequences are analysed using image processing, in order to obtain actual traffic area. Then, time for green light is allocated according to that traffic area. Hence, wastage of effective green time is eliminated by the proposed method since it allocates time to green signal that is sufficient to pass the actual traffic presented on the road. Results reveals, effective green time that need to pass the traffic, is proportional to the road area covered by traffic at that time.