Research Papers - Dept of Information Technology
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/593
Browse
405 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Embargo Data-centric single teacher guided knowledge distillation for alleviating sub-optimal supervision in image classification(Elsevier Ltd, 2026-02-23) Sharma, K; Silva, B. NIn recent years, larger, deeper, and more complex deep learning models have emerged as a result of advancements in deep learning techniques. Nevertheless, the computational costs have also increased with the growing model size. Thus, Knowledge Distillation has evolved into a cornerstone in contemporary machine learning, facilitating the transfer of knowledge from cumbersome teacher models to more compact student models. However, student learning is persistently challenged by sub-optimal supervision caused by erroneous and ambiguous teacher predictions. Moreover, the learning process is further deteriorated by the complications introduced through frequently encountered noisy labels in real-world datasets. Existing methods often resort to the ensemble of teachers, introducing additional complexity. We propose a novel, simple, and efficient learning method, Corrective Knowledge Distillation (CKD), to alleviate these drawbacks while relying solely on a single-teacher model. The proposed work employs a two-phase learning paradigm. In the initial phase, the teacher selectively teaches extremely confident knowledge to the student, and in the subsequent phase, the student leverages its own past learning experiences, conditioning its knowledge acquisition on the guidance of the teacher. The proposed method consistently exhibits superior performance in addressing sub-optimal supervision, as evidenced by comprehensive experiments on benchmark datasets such as CIFAR-100, CIFAR-100N-Fine, and ImageNet-1K. Notably, CKD surpasses established baselines, achieving substantial accuracy gains of up to 3.53% in real-world scenarios. Furthermore, CKD exhibits exceptional robustness in highly noisy environments, outperforming ensemble techniques by a significant margin of up to 5.18%. Our code is available at https://github.com/Karthick47v2/ckd.Publication Open Access Adding Common Sense to Robots Using Ontology(International Association of Computer Science and Information Technology, 2025-04-11) Ranathunga, R.A.A.L; Rajapaksha, SThis work investigates how ontological frameworks might improve robots’ ability to reason using common sense. The goal of the project was to enhance robot decision-making in dynamic real-world situations by developing an ontology-based model retraining technique. The researchers wanted to incorporate organized commonsense knowledge into robotic systems, so they built extensive ontologies that captured knowledge about the physical world and human interactions. The research compared the performance of robots with conventional models (control group) to those with ontology-enhanced models (experimental group) across various measures. The results indicate that this strategy may be used to develop more competent and user-friendly robotic helpers for a variety of sectors, including industry, healthcare, and education. Although the study has limitations related to data quality and experimental design, it does demonstrate the promise of ontology-based techniques to advance autonomous systems and human-robot interactions. Extending ontology databases, multidisciplinary cooperation, and investigating applications in other sectors are some of the future research goals.Publication Open Access Bridging Language Barriers in Programming Education: Java Programming Assistance Tool for Sinhala Native Speakers(International Association of Computer Science and Information Technology, 2025-09-12) Athukorala, K. S.N; De Silva, D.IThis study presents an innovative programming assistance tool designed to address language barriers faced by Sinhala-speaking novice Java programmers. The tool provides real-time Java code generation and diagram creation based on Sinhala programming queries, enhancing conceptual understanding. Developed using a Design-Based Research methodology, the tool underwent iterative testing with 122 Sinhala-speaking learners, incorporating user feedback to refine usability and performance. Central to the system is Generative Pre-trained Transformer, version 3.5 Turbo, ensuring accurate translations and programming assistance, alongside a transformer-based model that translates Sinhala queries into English for processing. The translation model achieved 91.37% accuracy, with strong Bilingual Evaluation Understudy scores validating its contextual relevance. The tool’s practical applications extend beyond academia, supporting educational institutions, self-learners, and industry professionals in learning and skill development. Statistical evaluation of user performance demonstrated significant improvements in programming comprehension, reinforcing its effectiveness. By promoting inclusivity and expanding access to programming knowledge, this research contributes to the advancement of Sri Lanka’s technology sector and establishes a scalable framework for broader implementation in multilingual programming education. CopyrightPublication Open Access The Potential of Arboreal Tiger Beetle (Derocrania scitiscabra Walker) as a Natural Enemy for the Control of Fall Armyworm (Spodoptera frugiperda JE Smith)(Faculty of Agricultural Sciences, Sabaragamuwa University of Sri Lanka, 2025-01-15) Kasige R.H.; Abeywardhana D.L; Pallewatta N; Perera M.T.M.D.R; Dangalle C.DPurpose: The fall armyworm, Spodoptera frugiperda, is a serious pest of maize, sugarcane, and other crops in Sri Lanka. Natural predators and parasitoids are considered the best methods to control this insect pest. The present study investigates the possibility of using an endemic tiger beetle, Derocrania scitiscabra, as a natural predator for the larval stages of S. frugiperda. Research Method: The feeding preferences of D. scitiscabra to live prey versus dead prey, live prey types including different instar stages of S. frugiperda were investigated in the laboratory using choice tests. Fresh minced meat was used as dead prey, while red ants, earthworms and the six larval instar stages of S. frugiperda were used as live prey. Findings: Red ants were the most preferred prey type of D. scitiscabra, and dead prey, earthworms and mature S. frugiperda larvae were not consumed. Early larval instar stages of S. frugiperda were selected as prey, and the beetle showed a high feeding preference for the second larval instar stage. This feeding preference was observed irrespective of whether red ants were present or absent in the same environment. D. scitiscabra may have selected the second larval instars of S. frugiperda due to their small size, high prey density, mobility, and being devoid of injury. Research Limitations: The tests were conducted under laboratory conditions in insectary facilities. However, field investigations are essential to understand the ecological dynamics that affect insect behavior and survival. Originality/value: An endemic beetle is introduced to control S. frugiperda infestations in their early stages of development. The finding may provide an environmentally safe and economically beneficial method to control S. frugiperda.Publication Embargo Deep Q-Network-Based Path Planning in a Simulated Warehouse Environment with SLAM Map Integration and Dynamic Obstacles(Department of Agribusiness, Universitas Muhammadiyah Yogyakarta, 2025-09-19) Medagangoda, H; Jayawickrama, N; de Silva, R; Samantha K.R.U.U; Abeygunawardhana, P.K.WWith the rise of e-Commerce and the evolution of robotic technologies, the focus on autonomous navigation within warehouse environments has increased. This study presents a simulation-based framework for path planning using Deep Q-Networks (DQN) in a warehouse environment modeled with moving obstacles. The proposed solution integrates a prebuilt map of the environment generated using Simultaneous Localization and Mapping (SLAM), which provides prior spatial knowledge of static obstacles. The reinforcement learning model is formulated with a state space derived from grayscale images that combine the static map generated by SLAM and dynamic obstacles in real time. The action space consists of four discrete movements for the agent. A reward shaping strategy includes a distance-based reward and penalty for collisions to encourage goal-reaching and discourage collisions. An epsilon-greedy policy with exponential decay is used to balance exploration and exploitation. This system was implemented in the Robot Operating System (ROS) and Gazebo simulation environment. The agent was trained over 1000 episodes and metrics such as the number of actions executed to reach the goal and the cumulative reward per episode were analyzed to evaluate the convergence of the proposed solution. The results across two goal locations show that incorporating the SLAM map enhances learning stability, with the agent reaching a goal approximately 150 times, nearly double the success rate compared to the baseline without map information, which achieved only 80 successful episodes over the same number of episodes. This indicates faster convergence and reduced exploration overhead due to improved spatial awareness.Publication Open Access Improved Path Planning for Multi-Robot Systems Using a Hybrid Probabilistic Roadmap and Genetic Algorithm Approach(Department of Agribusiness, Universitas Muhammadiyah Yogyakarta, 2025-03-24) Jathunga, T; Rajapaksha, SThis study focuses on the development and application of an improved Probabilistic Roadmap (PRM) algorithm enhanced with Genetic Algorithms (GA) for multi-robot path planning in dynamic environments. Traditional PRM-based methods often struggle with optimizing path length and minimizing turns, particularly in complex, multi-agent scenarios. To address these limitations, we propose a hybrid PRM-GA approach that incorporates genetic operators to evolve optimal paths for multiple robots in real-time.The research contribution is an enhanced PRM-GA framework that improves efficiency in multi-robot navigation by integrating evolutionary techniques for dynamic obstacle handling and optimized path generation.The research methodology involves testing the algorithm in various environments, including varying robot numbers and environmental complexities, to evaluate its scalability and effectiveness. Our results demonstrate that the PRM-GA algorithm successfully reduces both path lengths and turn counts compared to standard PRM-based methods, ensuring collision-free and smooth paths. The algorithm showed robust performance across different scenarios, effectively handling dynamic obstacles and multi-agent coordination. However, in highly dynamic environments with rapidly changing obstacles and constraints, the algorithm may occasionally produce paths with turn counts and distances similar to or slightly higher than those of simpler approaches due to the need for frequent re-optimization. Future research can explore incorporating additional factors such as energy consumption and time optimization, alongside distance and turns, to further enhance the algorithm's efficiency in real-world applications. Overall, the PRM-GA approach advances the state of the art by offering a more adaptable and scalable solution for multi-robot path planning, with applications in logistics, industrial automation, and autonomous robotics.Publication Open Access Comparison of cardiovascular risk prediction models developed using machine learning based on data from a Sri Lankan cohort with World Health Organization risk charts for predicting cardiovascular risk among Sri Lankans: A cohort study(BMJ Publishing Group, 2025-01-15) Mettananda, C; Solangaarachchige, M; Haddela, P; Dassanayake, A.S; Kasturiratne, A; Wickremasinghe, R; Kato, N; De Silva, H.JIntroduction Models derived from non-Sri Lankan cohorts are used for cardiovascular (CV) risk stratification of Sri Lankans. Objective To develop a CV risk prediction model using machine learning (ML) based on data from a Sri Lankan cohort followed up for 10 years, and to compare the predictions with WHO risk charts. Design Cohort study. Setting The Ragama Health Study (RHS), an ongoing, prospective, population-based cohort study of patients randomly selected from the Ragama Medical Office of Heath area, Sri Lanka, focusing on the epidemiology of non-communicable diseases, was used to develop the model. The external validation cohort included patients admitted to Colombo North Teaching Hospital (CNTH), a tertiary care hospital in Sri Lanka, from January 2019 through August 2020. Participants All RHS participants, aged 40-64 years in 2007, without cardiovascular disease (CVD) at baseline, who had complete data of 10-year outcome by 2017, were used for model development. Patients aged 40-74 years admitted to CNTH during the study period with incident CV events or a disease other than an acute CV event (CVE) with complete data for CVD risk calculation were used for external validation of the model. Methods Using the follow-up data of the cohort, we developed two ML models for predicting 10-year CV risk using six conventional CV risk variables (age, gender, smoking status, systolic blood pressure, history of diabetes, and total cholesterol level) and all available variables (n=75). The ML models were derived using classification algorithms of the supervised learning technique. We compared the predictive performance of our ML models with WHO risk charts (2019, Southeast Asia) using area under the receiver operating characteristic curves (AUC-ROC) and calibration plots. We validated the 6-variable model in an external hospital-based cohort. Results Of the 2596 participants in the baseline cohort, 179 incident CVEs were observed over 10 years. WHO risk charts predicted only 10 CVEs (AUC-ROC: 0.51, 95% CI 0.42 to 0.60), while the new 6-variable ML model predicted 125 CVEs (AUC-ROC: 0.72, 95% CI 0.66 to 0.78) and the 75-variable ML model predicted 124 CVEs (AUC-ROC: 0.74, 95% CI 0.68 to 0.80). Calibration results (Hosmer-Lemeshow test) for the 6-variable ML model and the WHO risk charts were χ 2 =12.85 (p=0.12) and χ 2 =15.58 (p=0.05), respectively. In the external validation cohort, the sensitivity, specificity, positive predictive value, negative predictive value, and calibration of the 6-variable ML model and the WHO risk charts, respectively, were: 70.3%, 94.9%, 87.3%, 86.6%, χ 2 =8.22, p=0.41 and 23.7%, 79.0%, 35.8%, 67.7%, χ 2 =81.94, p<0.0001. Conclusions ML-based models derived from a cohort of Sri Lankans improved the overall accuracy of CV-risk prediction compared with the WHO risk charts for this cohort of Southeast Asians.Publication Embargo Eco-friendly bismuth halide chalcogenide perovskites for solar energy harvesting(Royal Society of Chemistry, 2025-03-04) Don Muditha Akmal, U. K; Hu, D; Wijesekara Abeygunawardhana, P.K; Sewvandi, G. AThe quest to eliminate lead (Pb) content in perovskite photovoltaic materials has significantly shifted focus towards identifying viable Pb-free alternatives. This study provides a comprehensive theoretical investigation of CH3NH3BiI2Se and CH3NH3BiI2S as Pb alternative candidates. Density Functional Theory (DFT) calculations and the solar cell capacitance simulator (SCAPS) were used. The DFT analysis reveals that both CH3NH3BiI2Se and CH3NH3BiI2S possess indirect band gaps of 1.35 eV and 1.39 eV, respectively. CH3NH3BiI2Se demonstrates a higher absorption coefficient, stronger absorption in the UV-visible regions, a broader absorption spectrum and better charge carrier mobilities compared to CH3NH3BiI2S. CH3NH3BiI2Se and CH3NH3BiI2S based solar cells which show 24.06% and 21.85% power conversion efficiencies (PCEs), respectively. This study emphasizes the potential of CH3NH3BiI2Se as a promising bismuth mixed halide chalcogenide compound for the development of sustainable perovskite solar cells. The findings provide a foundation for the guided design of novel bismuth chalcogenide compounds for optoelectronic applications and experimental studies.Publication Open Access Real-Time Coordinate Estimation for SCARA Robots in PCB Repair Using Vision and Laser Triangulation(Multidisciplinary Digital Publishing Institute (MDPI), 2025-04-07) Sanjeewa, N; Wathudura, V. M; Kahatapitiya, N. S; Silva, B. N; Subasinghage, K.; Wijesinghe, R.EThe Printed Circuit Board (PCB) manufacturing industry is a rapidly expanding sector, fueled by advanced technologies and precision-oriented production processes. The placement of Surface-Mount Device (SMD) components in PCB assembly is efficiently automated using robots and design software-generated coordinate files; however, the PCB repair process remains significantly more complex and challenging. Repairing faulty PCBs, particularly replacing defective SMD components, requires high precision and significant manual expertise, making automated solutions both rare and difficult to implement. This study introduces a novel real-time machine vision-based coordinate estimation system designed for estimating the coordinates of SMD components during soldering or desoldering tasks. The system was specifically designed for Selective Compliance Articulated Robot Arm (SCARA) robots to overcome the challenges of repairing miniature PCB components. The proposed system integrates Image-Based Visual Servoing (IBVS) for precise X and Y coordinate estimation and a simplified laser triangulation method for Z-axis depth estimation. The system demonstrated accuracy rates of 98% for X and Y axes and 99% for the Z axis, coupled with high operational speed. The developed solution highlights the potential for automating PCB repair processes by enabling SCARA robots to execute precise picking and placement tasks. When equipped with a hot-air gun as the end-effector, the system could enable automated soldering and desoldering, effectively replacing faulty SMD components without human intervention. This advancement has the potential to bridge a critical gap in the PCB repair industry, improving efficiency and reducing dependence on manual expertise.Publication Embargo Tuning of Optoelectronic Properties of Chalcohalides by Tailoring Pnictogen Composition for Sustainable Photovoltaics(John Wiley and Sons Inc, 2025-08) Hu, D; Abeygunawardhana, P.K.W; Asha, GThis study investigates Sb1-xBixSeI pnictogen chalcohalides as lead-free materials for photovoltaic and optoelectronic applications using density functional theory (DFT) calculations. Increasing Bi content from 0.5 to 0.6 reduces the bandgap from 1.60 to 1.43 eV, enhancing the light absorption and aligning with the optimal range for solar energy conversions. Structural analysis reveals that higher Bi substitution expands the lattice, reduces the hole effective mass, and improves the hole mobility, while the electron mobility decreases slightly. Sb0.4Bi0.6SeI demonstrates quasi-direct bandgap characteristics attributed to Bi-induced lattice distortion and strong spin–orbit coupling (SOC), which reduces the conduction band minimum and facilitate direct-like electronic transitions. Enhanced absorption near the band edge and localized states contribute to higher sub-bandgap absorption, broadening the spectral response. Reduced bandgap falls within the optimal range for single-junction solar cells, increasing photocurrent generation. While defect-induced recombination poses challenges, passivation and compositional tuning can optimize its performance. This study identifies the potential of Sb0.4Bi0.6SeI as a versatile absorber material in emerging solar cell architectures. The findings provide a pathway toward designing cost-effective and sustainable materials with tailored properties for next-generation photovoltaic and optoelectronic technologies.