Research Papers - Dept of Computer Systems Engineering

Permanent URI for this collection https://rda.sliit.lk/handle/123456789/1253

Browse

Filters

2014 - 201932020 - 20212

Advanced Search

Filter by

Settings

Author: search.filters.author.Dharmasena, T

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    PublicationEmbargo
    Image Enhancement Algorithm for Remote Surveillance Using Low Cost Cameras in Unevenly Illuminated Environments
    (IEEE, 2021-12-07) Dharmasena, T; Abeygunawardhana, P. K. W
    The surveillance domain is becoming increasingly popular due to the demand for security and safety. In terms of the data being used, the cameras linked to them play an essential role. However, in poorly lit environments, the acquired image may appear to be underexposed when the camera adjusts itself to the light available in the scene, resulting in a loss of visual information in poorly lit regions. This could result in poor functionality in image based processing, as well as a negative impact on remote surveillance monitoring activities. This study offers a novel algorithm for improving the illumination of scenes that are unevenly lit using a selective pixels processing method to enhance the illumination of the environment. Rather than processing grayscale versions of red,green and blue channels to enhance the image, this approach uses value (v) channel of an HSV image.
  • Thumbnail Image
    PublicationOpen Access
    Analysis and improvement of a construction permit approval process: A teaching case for developing business process development capabilities, targeting developing nations
    (ACIS, 2014) Weerasinghe, K; Bandara, W; Dharmasena, T; Kuruppubandara, M; Nawinna, D. P
    With the increasing competitiveness in global markets, many developing nations are striving to constantly improve their services in search for the next competitive edge. As a result, the demand and need for Business Process Management (BPM) in these regions is seeing a rapid rise. Yet there exists a lack of professional expertise and knowledge to cater to that need. Therefore, the development of well-structured BPM training/ education programs has become an urgent requirement for these industries. Furthermore, the lack of textbooks or other self-educating material, that go beyond the basics of BPM, further ratifies the need for case based teaching and related cases that enable the next generation of professionals in these countries. Teaching cases create an authentic learning environment where complexities and challenges of the ‘real world’ can be presented in a narrative, enabling students to evolve crucial skills such as problem analysis, problem solving, creativity within constraints as well as the application of appropriate tools (BPMN) and techniques (including best practices and benchmarking) within richer and real scenarios. The aim of this paper is to provide a comprehensive teaching case demonstrating the means to tackle any developing nation’s legacy government process undermined by inefficiency and ineffectiveness. The paper also includes thorough teaching notes The article is presented in three main parts: (i) Introduction - that provides a brief background setting the context of this paper, (ii) The Teaching Case, and (iii) Teaching notes.
  • Thumbnail Image
    PublicationEmbargo
    Digital Preservation and Noise Reduction using Machine Learning
    (IEEE, 2021-12-09) Aravinda, K. P; Sandeepa, K. G. H; Sedara, V. V; Chamodya, A. K. Y. L; Dharmasena, T; Abeygunawardhana, P. K. W
    This paper proposes a digital preservation solution for Sinhala audios to conserve those as documents with noise reduction. The solution has implemented multiple noise reduction techniques as a pre-processing step to remove unwanted internal and external noises. A two-step, two-way noise reduction process is applied to produce clean audios based on Deep Convolutional Neural Network (DCNN) and adaptive filter-based techniques. This approach implements two separate noise reduction models for internal and external noises. After that, the speech recognition decoder recognizes the speech and converts it to a Unicode document by acoustic, language, and pronunciation models using extracted audio features from the denoised audio. Further, noise reduction models are decoupled from the preservation solution and exposed as a sub solution for multilingualism noise reduction, supporting English and Sinhala audios.
  • Thumbnail Image
    PublicationEmbargo
    Design and implementation of an autonomous indoor surveillance robot based on raspberry pi
    (IEEE, 2019-12-05) Dharmasena, T; Abeygunawardhana, P. K. W
    In recent years robotics has influenced many fields including the security and surveillance domain. Due to convenience and flexibility more and more security robots tend to be deployed in place of humans for routine activities such as area sweeps. While there are many kinds of research have been done regarding this concern, many of the solutions cost more due to their implementation complexity while low-cost implementations are only capable of doing simple activities such as following given local coordinates. This paper describes an autonomous surveillance robot that is being developed while keeping the development costs as low as possible and is capable of performing routine patrols autonomously in indoor environments and detect anomalies around it such as temperature fluctuations, unauthorized personals and report them back to a central computer. This robot can be controlled remotely by security personals to facilitate manual inspections. Due to the development architecture of the robot, more software-based features can be added easily.
  • Thumbnail Image
    PublicationEmbargo
    Autonomous cloud robotic system for smart agriculture
    (IEEE, 2019-07-03) Dharmasena, T; De Silva, R; Abhayasingha, N; Abeygunawardhana, P. W. K
    Agriculture sector occupies 25.9% of the world employment. The demand for food production is rapidly increasing with the increase of world population. Developing the existing agricultural infrastructure by incorporating modern technologies will help to match this increasing demand. This paper proposes a automated system to optimally control the climate and irrigation in a greenhouse by monitoring temperature, soil moisture, humidity and pH through a cloud connected mobile robot which can detect the unhealthy plants using image processing. A fuzzy controller will control the heating and cooling system, irrigation system and humidifiers installed in the greenhouse based on the sensor readings. The mobile robot navigates through a predefined map of the greenhouse and collect soil samples to perform measurements while onboard sensors will collect the ambient climate data. A camera mounted on the mobile robot will capture the plant and detect unhealthy crops based on the colour and the texture of the leaves.