http://repository.ou.ac.lk/handle/123456789/4269 Fri, 09 Oct 2026 16:14:46 GMT 2026-10-09T16:14:46Z http://repository.ou.ac.lk/handle/123456789/4291 A Review of Flood Hazard Assessment and Risk Zone Identification Using Conventional Physical Models and Machine Learning Algorithms: A Study of the Kalu River Basin Kisho Raj, Uthayashankar; Iresh, Shahika; Athapattu, Bandunee The Kalu River Basin in southwestern Sri Lanka has experienced significant flooding, primarily from monsoonal and cyclonic rainfall. Steep topography, rapid land-use changes, and climate change exacerbate flood risk. Notable flood events in this region occurred in 2003, 2008, 2012, 2014, 2017, 2018, and 2021, causing substantial economic losses and social impacts. Traditional forecasting methods have relied on physically based hydrological and hydrodynamic models to simulate rainfall–runoff relationships, particularly in well-instrumented catchments. However, these conventional models often require extensive data, presenting challenges in data-limited environments such as Sri Lanka. Machine learning (ML) models offer an alternative by learning flood-generating patterns directly from hydrological and environmental data, achieving high forecasting accuracy with limited information. In contrast, process-based models encounter challenges such as inflexibility, nonlinear dynamics, and residual biases. This paper reviews global and regional research focusing on hybrid approaches that integrate physical models with ML techniques and assesses their potential application in the Kalu River Basin. The review emphasizes the significance of combining physical and ML models, demonstrating that hybrid methods generally outperform standalone models in estimating peak flows and predicting extreme events, even under data-limited conditions. In summary, this review advocates adopting integrated modeling systems as robust and adaptable solutions for flood forecasting. Thu, 01 Jan 2026 00:00:00 GMT http://repository.ou.ac.lk/handle/123456789/4291 2026-01-01T00:00:00Z http://repository.ou.ac.lk/handle/123456789/4290 Sri Lankan Tamil Sign Language Recognition Using Machine Learning and Computer Vision: A Systematic Literature Review Kanesanathan, Ealisai; Kirushika, J. Sign language recognition has become an important research area in assistive technology, with machine learning and computer vision providing opportunities to support communication between deaf or hard-of-hearing people and the wider community. Existing studies have investigated different approaches for recognizing sign language gestures from images and video, including convolutional neural networks (CNNs), long short-term memory (LSTM) networks, support vector machines (SVMs), wearable devices, depth cameras, and mobile-based systems. However, much of the existing research focuses on widely studied sign languages or Sinhala Sign Language, while research specifically related to Sri Lankan Tamil Sign Language (TSL) remains limited. Therefore, this systematic literature review examines existing machine learning and computer vision approaches for sign language recognition and translation, with particular attention to their relevance to TSL. A total of 18 studies were reviewed based on the sign language considered, type of input data, recognition task, machine learning technique, dataset, translation approach, hardware requirements, and reported limitations. The review shows that CNN-based approaches are commonly used for visual gesture recognition, while LSTM and other sequence-based approaches have been applied to continuous sign recognition. SVM- based methods and specialized hardware such as depth cameras and wearable devices have also been reported. However, the reviewed studies indicate limited availability of TSL-specific datasets and limited research on continuous TSL recognition, Tamil language translation, and low-cost real-time implementation. These findings indicate the need for further TSL-specific research, particularly the development of suitable datasets and the evaluation of computer vision and machine learning approaches using TSL data. The findings of this review provide a basis for future research on practical TSL recognition and translation systems using affordable camera-based devices. Thu, 01 Jan 2026 00:00:00 GMT http://repository.ou.ac.lk/handle/123456789/4290 2026-01-01T00:00:00Z http://repository.ou.ac.lk/handle/123456789/4289 MyAiPlate: A Mobile Application for Personalized Nutrition Using YOLOv8 Food Recognition and a Hybrid LSTM-DQN Model for Real-Time Meal Plan Adaptation Zaina, M.Z.; Nusri, M.A.A.; Thafani, M.N.; Sharafa, H.M.F.; Nasheeth, M.B.N.M. Keeping a balanced and individualized diet plays an important role in the prevention of chronic diseases as well as in maintaining optimal wellness. The accurate monitoring of daily calorie consumption and abiding by the restrictions of certain diets is key to achieving good health. Nevertheless, existing dieting apps are not dynamic enough; they only serve as online diaries incapable of adjusting according to the behaviour of the user. A multi-model machine learning architecture was developed for the MyAiPlate application to address this gap. A YOLOv8 object detection model, fine-tuned on a real-world dataset of 1,169 labelled food images, was implemented to visually recognize food items and estimate caloric intake from user-captured images. The proposed machine learning components were rigorously evaluated for performance and user safety. The YOLOv8 food recognition model was validated using unseen, real-world meal images. On the held-out validation set the model achieved 95.16% precision, 94.42% recall, 97.51% mAP@50, and 75.57% mAP@50-95. In addition to encouraging user participation, feedback was provided through a conversational interface within the application. Traditional diet tracking tools do not have a conversational approach and cannot answer specific user inquiries in a context-aware manner. An AI Nutrition Chatbot was implemented using Large Language Models (LLMs) based on transformers from OpenRouter, which offered real-time, context-aware, personalized nutritional advice based on meal entries. The conversational interface was tested using 50 queries, including 25 challenging nutritional queries and 25 common diet prompts to ensure that the LLM could deliver reliable results without hallucinations. Thu, 01 Jan 2026 00:00:00 GMT http://repository.ou.ac.lk/handle/123456789/4289 2026-01-01T00:00:00Z http://repository.ou.ac.lk/handle/123456789/4288 INTEGRATING ARTIFICIAL INTELLIGENCE AND SATELLITE TELEMETRY INTO COASTAL CONSERVATION AND MARINE POLLUTION LEGISLATION: A FEASIBLE LEGISLATIVE MODEL FOR SRI LANKA De. Zoysa, Sasith The vulnerability of island nations to catastrophic marine pollution and severe coastal degradation - exemplified by the MV X-Press Pearl disaster in Sri Lankan waters - reveals fundamental structural limitations in traditional environmental governance. Sri Lanka’s primary legislative frameworks, the Coast Conservation and Coastal Resource Management Act No. 57 of 1981 (as amended) and the Marine Pollution Prevention Act No. 35 of 2008, depend primarily on post- hoc reactive physical inspections and static, periodic planning cycles. This study formulates a feasible legal model that integrates Artificial Intelligence (AI), Synthetic Aperture Radar (SAR) telemetry, multispectral optical remote sensing, and automated data fusion into Sri Lanka’s statutory framework. Through an analytical examination of domestic legislation, international conventions (UNCLOS, MARPOL 73/78, Basel Convention), and recent environmental jurisprudence, this paper identifies critical statutory lacunae regarding data mandates, cross-border intelligence exchange, and the evidentiary admissibility of autonomous algorithmic detections. To resolve these deficiencies, this study proposes two statutory draft amendments: (1) Section 12A of Act No. 57 of 1981, mandating an automated, real- time "Smart Coastal Zone Management System" (Smart- CZMP); and (2) Section 25A of Act No. 35 of 2008, establishing statutory rebuttable presumptions of liability derived from validated AI satellite surveillance alongside procedural compliance with the Evidence (Special Provisions) Act No. 14 of 1995. This legal framework provides a proactive, technologically sovereign mechanism to mitigate marine plastic accumulation, coastal erosion, and maritime environmental crimes. Thu, 01 Jan 2026 00:00:00 GMT http://repository.ou.ac.lk/handle/123456789/4288 2026-01-01T00:00:00Z