Browsing by Author "Jayamanne, D."
Now showing 1 - 5 of 5
Results Per Page
Sort Options
- Item3D Walk-through at NSBM(NSBM Green University, 2019-11-22) Manujaya, D.; Jayamanne, D.In this work, a 3D Virtual Walk through of NSBM using the Unity Game Engine is presented. The walk-through includes real life and detailed 3D models of buildings and objects. The buildings are modelled using SketchUp including the interiors. All buildings and objects are approximately modelled to scale. Objects are also modelled using SketchUp where possible. All other models which require more details are modelled using 3Ds Max and Blender. As the walk-through is started, a user is dropped in front of a building. Then, the user can control the camera using the mouse. The arrow keys are used to move forward, back, left and right in the virtual environment. Also, the user is able to navigate inside the buildings, walk into rooms and open doors. The light source above the ground that imitates the sun provides lighting and shadows for the buildings and trees. There are separate light sources inside each room to model the actual lighting conditions we observe in real life. All lighting is handled using the Unity Engine. In this work, the user has the freedom to move the camera wherever the user desires in the given area as there are no fixed cameras, predefined paths, and environmental hotspots. However, the user is able access any part of the model if it is physically possible.
- ItemFair Price Prediction System for Used Cars in Sri Lanka Using Machine Learning and Robotic Process Automation(NSBM Green University, 2019-11-22) Jayadeera, T.P.; Jayamanne, D.The prices of most of the brand-new cars are rapidly increased by car manufacturers due to the increment of prices in raw materials and inflation rates. Also, the taxes for importing brand new cars have been rising consistently during the last decade in Sri Lanka. Due to these factors, most of the time Sri Lankan middle and lower-class people tend to buy used cars (Toyota Corolla) rather importing brand-new cars. With this increment of demand in used cars, some of the used car sellers take advantage of this scenario by listing unrealistic prices for the used cars and most of the used car buyers getting caught on this. This work focuses on creating a used car price prediction system for Sri Lanka using supervised learning techniques. For the study, Analysis with different suitable machine learning models is performed using an online data set to discover the best suitable regression models. The selected models are trained using an actual Sri Lankan used car data set which is expected to be extracted from online car advertisement websites with the help of Robotic Process Automation (RPA) technology. In the end, a web application will be presented to the users (Used car buyers) which accepts features of the car as inputs and provide a predicted fair price for the expected particular used car as the output.
- ItemHealthXStream: A Privacy Focused Doctor Meetup Solution for the General Public(NSBM Green University, 2018-08-25) Dissanayake, S.; Jayamanne, D.“HealthXStream” is a privacy oriented, fully secure, health monitoring platform where patients and doctors get connected with each other to solve non-severe health problems. There are very few software applications in the world providing consumer grade doctor meetups through the mobile phone. Even from that few, “HealthXStream” being unique as it provides the real-time pulse rate and the temperature while providing the privacy at its peak. And it is the only health monitoring application with privacy in mind which complies to the European legislations. Doctors and the Patients are the main actors where the patients could chat with the doctor, while the pulse rate/body temperature streamed. This solution is mainly targeted to people in senility and for the people who are not preferring to move outside to meet the doctor. This will address the issue of inconvenience in accessing the doctor in a senility environment where privacy and security matters.
- ItemPoster Presentation: Video Based Vehicle Tracking and Monitoring System(NSBM Green University, 2018-08-25) Igala Guruge, V.; Jayamanne, D.Computer vision technology has paved the way to process video based traffic monitoring as an alternative way to traditional monitoring methods. In vehicle traffic monitoring, there are three main tasks namely vehicle detection, tracking and classification. This paper presents a feasible solution for vehicle detection, tacking and classification from a video via extracting and analyzing video frames. The background is modeled using a background model and the foreground is estimated using background subtraction. Then after applying morphological operations, the presence of a vehicle is detected on the foreground and classified according to its type. The system is also able to count the number of vehicles with in a defined region of interest.
- ItemVehicle License Plate Recognition System(NSBM Green University, 2018-08-25) Jayamanne, D.This paper presents a method for automatic detection and recognition of vehicle license plates in a Sri Lankan context. License Plates of the still images capturing the frontal view of vehicles are initially located by identifying the regions with the highest vertical edge density. After applying dilation and connected component analysis, the license plate can be localized. The Hough Transformation is used on the extracted plate region to find the skew angle which is used to rotate the plate region to align it with the horizontal axis. Then, the intensity of the plate region is enhanced using morphological operations before obtaining the corresponding binary image. The plate area is further pre-processed before extracting the characters. Boundaries of the characters are identified by the row sum of pixels and column sum of pixels. Extracted characters are normalized to 16 x 10. Then, template matching is used to recognize the characters of the license plate. The developed license plate recognition system has an accuracy of 84% for detecting the number plates of the still images of vehicles.