Research Papers - Dept of Software Engineering
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/1022
Browse
226 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Open Access Optical Insight: Enhancing Ophthalmic Diagnostics with Automated Detection of Retinal Abnormalities(International Association of Computer Science and Information Technology, 2025-06-11) De Silva, D.I; Wijendra, D. R; Siriwardana, K.S; Gunasekara, S.N.W; Piyumantha, U; Thilakaratne, S.PEarly and accurate detection of retinal diseases is crucial for preventing vision loss, yet traditional diagnostic methods remain limited by subjectivity and inefficiencies. This study introduces an Artificial Intelligence (AI)-driven diagnostic system leveraging hybrid deep learning models to detect Glaucoma, Macular Hole, Central Serous Retinopathy, and Drusen using fundus images. By integrating multiple architectures, including Residual Network (ResNet), Visual Geometry Group 16-layer network (VGG16), Densely Connected Convolutional Network (DenseNet), U-shaped Network (U-Net), and You Only Look Once version 8 (extra-large variant) (YOLOv8x), the system enhances diagnostic precision and generalization across diverse imaging conditions. Key innovations include the hybrid ResNet-VGG16 and DenseNet-VGG16 models, which significantly improve detection accuracy for Drusen and Central Serous Retinopathy, respectively. Additionally, the U-Net-ResNet hybrid architecture mitigates overfitting, ensuring more reliable Macular Hole detection, while the YOLOv8x object detection model outperforms traditional approaches in Glaucoma localization by accurately identifying the optic disc. These models, integrated into a web-based diagnostic platform, achieved sensitivities and specificities exceeding 95%, establishing a new performance benchmark for automated ophthalmic diagnostics. This research advances medical image analysis by demonstrating the efficacy of hybrid deep learning models, offering a scalable AI solution for early retinal disease detection. Its integration into clinical workflows highlights its potential to transform ophthalmic care, enhancing accessibility and improving patient outcomes.Publication Embargo Spatio-temporal graph neural network based child action recognition using data-efficient methods: A systematic analysis(Elsevier Inc, 2025-06-03) Mohottala, S; Gawesha, A; Kasthurirathna, D; Samarasinghe, P; Abhayaratne, CThis paper presents implementations on child activity recognition (CAR) using spatial–temporal graph neural network (ST-GNN)-based deep learning models with the skeleton modality. Prior implementations in this domain have predominantly utilized CNN, LSTM, and other methods, despite the superior performance potential of graph neural networks. To the best of our knowledge, this study is the first to use an ST-GNN model for child activity recognition employing both in-the-lab, in-the-wild, and in-the-deployment skeleton data. To overcome the challenges posed by small publicly available child action datasets, transfer learning methods such as feature extraction and fine-tuning were applied to enhance model performance. As a principal contribution, we developed an ST-GNN-based skeleton modality model that, despite using a relatively small child action dataset, achieved superior performance (94.81%) compared to implementations trained on a significantly larger (x10) adult action dataset (90.6%) for a similar subset of actions. With ST-GCN-based feature extraction and fine-tuning methods, accuracy improved by 10%–40% compared to vanilla implementations, achieving a maximum accuracy of 94.81%. Additionally, implementations with other ST-GNN models demonstrated further accuracy improvements of 15%–45% over the ST-GCN baseline. The results on activity datasets empirically demonstrate that class diversity, dataset size, and careful selection of pre-training datasets significantly enhance accuracy. In-the-wild and in-the-deployment implementations confirm the real-world applicability of above approaches, with the ST-GNN model achieving 11 FPS on streaming data. Finally, preliminary evidence on the impact of graph expressivity and graph rewiring on accuracy of small dataset-based models is provided, outlining potential directions for future research. The codes are available at https://github.com/sankamohotttala/ST_GNN_HAR_DEML.Publication Embargo Revolutionalize Your Learning Experience with EQU ACCESS(IEEE, 2024-07-25) Raveenthiran, G; Sivarajah, K; Kugathasan, V; Chandrasiri, S; Mohamed Riyal, A. A; Rajendran, KThis paper introduces a novel approach aimed at enhancing online education by placing a central focus on students' emotional well-being and improving their learning experiences. The approach integrates four key machine learning technologies: behavioral expression analysis, a personalized chatbot for emotional support, voice stress detection, and visual content description. Through empirical findings, the study illustrates the effectiveness of these methods in bolstering students' emotional well-being and academic performance. By providing a roadmap for the advancement of online education and emotional support, this research holds promise for delivering substantial benefits to learners worldwide. The study showcases notable advancements in online education, reporting a 30% rise in perceived emotional support and a 25% increase in overall satisfaction. The personalized emotional support chatbot achieved an 85% accuracy in addressing students' emotional needs, while voice stress detection boasted a 90% accuracy in identifying anxiety. Additionally, visual content description led to a 20% improvement in comprehension. These findings highlight the approach's potential to elevate both emotional well-being and academic performance in online learners.Publication Embargo Advancing Canine Health and Care: A Multifaceted Approach using Machine Learning(IEEE, 2023-06-26) Wimukthi, Y; Kottegoda, H; Andaraweera, D; Palihena, P; Fernando, H; Kasthurirathnae, DThis research paper proposes a comprehensive approach to enhance the well-being of dogs through a range of innovative technologies. Firstly, we develop an automated system for dog breed and age identification using a Convolutional Neural Network (CNN) and a transfer learning model. This system aims to provide an efficient and reliable solution for dog owners and new adopters who are interested in discovering more about their canine companions. Secondly, we propose the development of a system that uses Reinforcement Learning to generate personalized meal plans based on a variety of factors such as the dog's breed, age, weight, health status, and emotional state. The system aims to provide dog owners with a reliable and effective tool for generating personalized meal plans that will enhance their pets' overall health and well-being. Thirdly, we present a dog disease recognition application that utilizes an artificial neural network (ANN) for identifying dog diseases based on their symptoms. Lastly, we introduce a real-time remote dog monitoring system using loT devices with edge computing to detect aggressive and anxious sounds. Our system provides an accurate classification of dog sounds related to aggression and anxiety, which can help dog owners detect and respond to potential issues early on. This research aims to provide dog owners and veterinarians with a range of technologies that can help them better understand and care for their furry friends.Publication Embargo IoT-Enabled Smart Solution for Rice Disease Detection, Yield Prediction, and Remediation(IEEE, 2023-06-26) Wanninayake, K.M.I.S; Bambaranda, L.G.S. W; Wickramaarachchi, T.I; Pathirana, U.C.S.L; Vidhanaarachchi, S; Nanayakkara, A.A.E.; Gunapala, K.R.D.; Sarathchandra, S.R.; Gamage, A.I; De Silva, D.ISri Lanka's rice cultivation is a vital industry supporting over 1.8 million cultivators and providing staple sustenance for 21.8 million people. According to Sri Lanka's Central Bank, rice cultivation contributed 2.7% to the country's GDP in 2020 [3]. Pests and diseases, particularly rice thrips damage and rice blast disease, are a challenge for the industry, as they cause yield loss. This paper describes an intelligent solution that aids stakeholders by detecting and classifying the disease, forecasting its dispersion, and providing remedies. The proposed solution is approached with deep learning techniques for real-time detection and classification of the disease, location tracking of infected areas, and pesticide application on the target. In addition, it predicts the spread of disease based on the locations of infected individuals. In addition, the solution enables Machine-learning algorithms to recommend appropriate rice varieties and predict yields. In controlled experiments utilizing data from Sri Lankan paddy fields, the proposed method obtained high accuracy rates of 89%-98% in identifying disease and rice varieties and yield prediction. This system has the potential to increase rice production and productivity, decrease yield loss, and benefit the Sri Lankan rice industry and producers.Publication Embargo DS-HPE: Deep Set for Head Pose Estimation(IEEE, 2023-04-18) Menan, V; Gawesha, A; Samarasinghe, p; Kasthurirathna, DHead pose estimation is a critical task that is fundamental to a variety of real-world applications, such as virtual and augmented reality, as well as human behavior analysis. In the past, facial landmark-based methods were the dominant approach to head pose estimation. However, recent research has demonstrated the effectiveness of landmark-free methods, which have achieved state-of-the-art (SOTA) results. In this study, we utilize the Deep Set architecture for the first time in the domain of head pose estimation. Deep Set is a specialized architecture that works on a “set” of data as a result of the “permutation-invariance” operator being utilized in the model. As a result, the model is a simple yet powerful and edge-computation-friendly method for estimating head pose. We evaluate our proposed method on two benchmark data sets, and we compare our method against SOTA methods on a challenging video-based data set. Our results indicate that our proposed method not only achieves comparable accuracy to these SOTA methods but also requires less computational time. Furthermore, the simplicity of our proposed method allows for its deployment in resource-constrained environments without the need for expensive hardware such as Graphics Processing Units (GPUs). This work underscores the importance of accurate and resource-efficient head pose estimation in the fields of computer vision and human-computer interaction, and the Deep Set architecture presents a promising approach to achieving this goal.Publication Embargo Interactive Mobile Application for Initial Skills Development of Primary Students in Sri Lanka(IEEE, 2022-12-09) Liyanage, C.; Kavinda, U. A. D. S.; Dasanayaka, D. S.; Shehara, P. G. J.; De Silva, D. I.In many cases, children between this age are using smartphones and other technology devices, to play games, watch cartoons, take photos and sometimes the chance is getting higher than we think that children access unnecessary contents due to lack of guidance and unawareness of parents. This interactive mobile application is used as an adaptive learning tool for the primary school students. Utilizing children’s comfort with technology allows for the development of their talents. In math skills development, some attractively designed gamified activities to solve basic math questions are given according to the skill level the child is currently in. The accuracy was much higher in the Convolutional Neural Network approach as it recorded a value of 0.9919. In environmental skills development component, the app will ask child to identify the surroundings according to a flow, starting from the house and towards the garden using object detection and the results were detected with a higher accuracy level around 0.9-0.99 after training the Machine Learning model. And in the language skills development component the child is given activities to develop pronunciation skills using audio processing and finally the verification of online achievements of a child by Non-Fungible Token technology, is fulfilled via the app.Publication Embargo AgroPro: Optimizer for Traditional Agricultural System in Sri Lanka(IEEE, 2022-12-09) De Silva, D.I.; Suriyawansa, G.M. T. K. D. S.; Senevirathna, M.R. U. M. T.; Balasuriya, I.D. I.; Deshapriya, A. G. S. P.; Gadiarachchi, G. A. D. K. M.Today, in many countries around the world, big data analysis and machine learning methods are used for industrial development. However, such techniques are rarely used in Sri Lankan agricultural industry. The success of agriculture depends heavily on the selection of the right crop. Choosing the right crop depends primarily on predicting future yields. Machine learning methods can be used very successfully to make future predictions about crop yields. Crop prediction mainly depends on the soil, geography, and climate of the growing location. Hence historical data with agricultural facts such as temperature, humidity, pH, and rainfall are used to predict yield as parameters in the machine learning function. Sri Lanka uses a traditional approach to distribute fertilizers among farmers. Not having an organized way to distribute fertilizers to the needed areas leads to many abnormalities along the way. As a result, the country is facing economic losses and resource wastage. Having an optimized distribution network is the key to overcoming those abnormalities. This research assesses the efficiency of the fertilizer distribution system and consists of time-series predictions on fertilizer usage to gain future value. The aim is to identify performance gaps in distribution management that lead to delayed fertilizer distribution affecting agricultural productivity.Publication Embargo Supervising Plant Growth in a Greenhouse(IEEE, 2022-11-30) Alexander, D.L; Hathnapitiya, G.A.G.; Gamage, A.I; Bandara, P.M.P.C; Giragama, G.W.M.N.U.I.B; Supunya Swamakantha, N.H.P.RSince the beginning of civilization, agriculture has played a significant part in the economy of a nation. Currently, as the population continues to increase at a rapid rate, arable lands are dwindling alongside urbanization. Even though farmers devote a substantial amount of time and effort to farming, environmental factors such as seasonal shifts can have a significant impact on the crop. Smart agriculture is implemented to boost the production of high-quality goods and address the lack of control over the farming process. The intelligent greenhouse technology proposed here is called “GSense,” and it could boost plant productivity by managing the greenhouse’s climate. In addition, this solution is useful for novices who are just beginning out in agriculture because it can make recommendations to its user. The execution of the solution is complemented by a mobile application and a desktop application via which the user may submit inputs and examine real-time sensor dataPublication Embargo Cocopal - A Deep Learning Based Intelligent System to Certify and Standardize the Quality of Coconut Based Products(IEEE, 2022-12-09) Gunawardana, K.H.R.; Deshan, M.P.N.; Hemachandra, M.G.S.P.; Ganegoda, D; Hettiarachchi, N. M; Weerasinghe, LThe procedure of certifying and standardizing the quality of the coconut-based products is done manually in Sri Lanka at precent. It is a time consuming and labor-intensive task and is conducted by experts. In most cases, the quality is decided solely by visual inspections by buyers and suppliers, with no scientific basis. The paper reports the capacity of bringing modern technology solutions such as Artificial Intelligence (AI), Machine Learning (ML), Image Processing (IP), and decentralized storage to aid in the certification and standardization of the quality of raw materials.Results showed that the accuracy of the proposed system is in the 86% to 90% range and showed that this technique could beimproved and used as an alternative to manual techniques.