Research Publications Authored by SLIIT Staff
Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4195
This collection includes all SLIIT staff publications presented at external conferences and published in external journals. The materials are organized by faculty to facilitate easy retrieval.
Browse
2522 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Open Access Early Diagnosis and Severity Assessment of Weligama Coconut Leaf Wilt Disease and Coconut Caterpillar Infestation Using Deep Learning-Based Image Processing Techniques(Institute of Electrical and Electronics Engineers Inc., 2025-02-07) Vidhanaarachchi, S; Wijekoon, Janaka L; Abeysiriwardhana, W. A. S. P; Wijesundara, Mlobal Coconut (Cocos nucifera (L.)) cultivation faces significant challenges, including yield loss, due to pest and disease outbreaks. In particular, Weligama Coconut Leaf Wilt Disease (WCWLD) and Coconut Caterpillar Infestation (CCI) damage coconut trees, causing severe coconut production loss in Sri Lanka and nearby coconut-producing countries. Currently, both WCWLD and CCI are detected through on-field human observations, a process that is not only time-consuming but also limits the early detection of infections. This paper presents a study conducted in Sri Lanka, demonstrating the effectiveness of employing transfer learning-based Convolutional Neural Network (CNN) and Mask Region-based-CNN (Mask R-CNN) to identify WCWLD and CCI at their early stages and to assess disease progression. Further, this paper presents the use of the You Only Look Once (YOLO) object detection model to count the number of caterpillars distributed on leaves with CCI. The introduced methods were tested and validated using datasets collected from Matara, Puttalam, and Makandura, Sri Lanka. The results show that the proposed methods identify WCWLD and CCI with an accuracy of 90% and 95%, respectively. In addition, the proposed WCWLD disease severity identification method classifies the severity with an accuracy of 97%. Furthermore, the accuracies of the object detection models for calculating the number of caterpillars in the leaflets were: YOLOv5-96.87%, YOLOv8-96.1%, and YOLO11-95.9%.Publication Embargo Rising from the crisis by raising their businesses exploring the social media strategies for micro-level women entrepreneur success(Emerald Publishing, 2025-10-03) Jayasuriya, N; Wisenthige, K; Kalansuriya, N; Munasinghe, S; Shamila, P; Swaris, SPurpose – This study examines the social media strategies adopted by micro-level women entrepreneurs who successfully transitioned their businesses from “brick and mortar” to online operations during the coronavirus (COVID-19) pandemic. Design/methodology/approach – The research consisted of two phases: semi-structured interviews followed by focus group discussions (FGDs). Twelve participants were selected through a judgmental sampling method for the interviews. The sample included contemporary successful micro-level female entrepreneurs who shifted to online platforms during the crisis. New factors were identified as determinants of success for online-based small businesses. The FGDs provided insights into these determinants, which were analyzed thematically. Findings – The findings indicate that the success of online businesses is influenced by an active presence on social media, the maintenance of a strong online network with customers and industry partners, vigilance in seizing strategic opportunities in a diverse environment, and the generation of powerful word-of-mouth (WOM) through excellent customer service. In contrast to gender inequality typically observed in business across Asian regions, no such gender stereotype issues were found among micro-level Sri Lankan women entrepreneurs. Originality/value – This study enhances the understanding of social media strategies in micro-entrepreneurship by identifying key success factors that have been underexplored in existing literature. It challenges assumptions about gender barriers in Asian entrepreneurship, demonstrating that Sri Lankan women entrepreneurs navigate digital business environments effectively. Unlike prior studies focused on larger firms or developed economies, this research highlights unique success determinants in an emerging market. The study introduces a conceptual framework that integrates social media engagement, strategic adaptability and network strength as critical components for online business success. These insights provide a foundation for future empirical studies in digital entrepreneurship.Publication Embargo Impact of Political Corruption on International Tourism Demand: A Panel Data Analysis of 22 Asian Countries(University of Nigeria Department of Mass Communication, 2026-06-01) Weerasinghe, P; Bandara, C; Indrapriya, A; Jayasekara, K; Yapa, S.R; Dayapathirana, NBackground: Tourism serves as a critical driver of international economic activity, particularly for developing nations that rely on the sector for employment generation and foreign exchange earnings. However, tourism performance varies significantly across jurisdictions, suggesting that institutional factors, such as political corruption, may play a decisive role in shaping industry outcomes. Objective: This study investigates the impact of political corruption on tourism demand across Asia. Specifically, it seeks to determine whether, and to what extent, corruption levels significantly deter international tourist arrivals in these countries. Methodology: Adopting a quantitative research design, the study utilises panel data analysis covering 22 Asian countries from 2002 to 2020. Simple linear regression models were employed to examine the relationship between political corruption indices and international tourism demand. Results: The findings reveal a statistically significant negative relationship between corruption and tourism demand. The regression analysis demonstrates that higher levels of perceived corruption are associated with a marked decrease in tourist inflows. Although the model exhibits relatively low explanatory power, the consistency and significance of the negative effect indicate that corruption is a critical institutional determinant of tourism performance. Conclusion: The study concludes that political corruption is a major deterrent to the growth of the tourism industry. Reducing institutional uncertainties and curbing corrupt practices are essential prerequisites for countries aiming to enhance their competitiveness and attract international visitors. Unique Contribution: This research contributes to the intersection of governance and tourism literature by providing long-term empirical evidence from the Asian context. It highlights the importance of institutional quality as a primary factor in the patterns of international tourism flows. Key Recommendation: Governments and policymakers in the Asian region should prioritise anti-corruption reforms as a strategic tool for tourism development. Strengthening institutional transparency and reducing political risks will create a more favourable environment for international travellers and foster sustainable economic growth.Item Open Access SLIF-Tomato: Inverted Residual Convolutional Block Attention Module for In-field Tomato Leaf Disease Recognition(Springer Science and Business Media Deutschland GmbH, 2026-06-27) George, R; Thuseethan, S; Ragel, R.G; Pakeerathan, K; Sivalingam, V; Mithuran, TTomato cultivation represents a critical component of nutrition, economic development, and public health, yet it is increasingly compromised by foliar diseases that diminish yield and intensify dependency on hazardous agrochemicals. Although deep learning models have demonstrated strong capabilities for automated disease recognition, existing benchmark datasets exhibit limited real-world utility, primarily due to the absence of in-field imagery and precise annotations of diseased regions. A novel attention mechanism, Inverted Residual Convolutional Block Attention Module (IR-CBAM), is proposed, combining inverted residual blocks with the CBAM Module, and is specifically tailored to address challenges posed by in-field image variability, such as complex backgrounds and inconsistent lighting. Furthermore, this study introduces SLIF-Tomato, the Sri Lankan In-Field Tomato leaf disease dataset, which is the first complete in-field dataset comprising class labels and bounding box annotations collected under diverse real-world conditions. The proposed approach achieved 99.66% and 99.91% accuracy rates on two curated versions of the SLIF-Tomato dataset. Subsequently, the YOLOv12-large model is employed to detect diseased regions, which obtained an average precision score of 88.5%. These contributions advance the development of accurate, efficient and field-adaptable diagnostic systems for tomato leaf disease management in precision agriculture.Publication Embargo Investigating the mechanical and durability properties of locally sourced fly ash-based ternary cement paste to increase OPC replacement(Emerald Publishing, 2026-07-07) Gamage, O; Kulasuriya, C; Perera, S V T JThe mechanical and durability performance of fly ash (FA)-based ternary blended cement pastes was investigated in an attempt to increase ordinary Portland cement (OPC) replacement while maintaining comparable strength and improved durability. This paper provides region-specific experimental data using locally sourced FA from Sri Lanka and imported ground granulated blast-furnace slag (GGBFS). A control mix (100% OPC), two binary blends and six ternary blends were tested for compressive strength at 3, 7 and 28 days, along with water absorption and resistance to sulfuric acid after 28 days. The results showed that OPC specimens achieved the highest early strength, while ternary blends with 25% and 40% OPC replacement achieved comparable 28-day strengths of 82.7 ± 4.8 MPa and 76.2 ± 3.2 MPa, respectively. GGBFS contributed to early strength gain, whereas FA enhanced long-term strength. Durability was improved significantly in the ternary blends. The F15S10 blend, with a higher GGBFS content, showed the greatest sulfuric acid resistance (4.8 ± 0.15% weight loss), while the F40S10 blend, with higher FA content, exhibited the lowest water absorption (6.0 ± 0.18%). Overall, ternary blends with 40% OPC replacement and low FA content are recommended for achieving balanced strength and superior durability in cement paste applications.Publication Open Access Chitosan-Starch Biocomposite for Enhanced Curcumin Delivery: Kinetic Modeling of pH and Ionic Strength Responsive Release and Evaluation of Biological Efficacy(American Chemical Society, 2026-07-21) Abeywickrama, L; Wijayawardana, S; Thambiliyagodage, C; Jayanetti, MA drug delivery system composed of chitosan and starch was used to deliver curcumin effectively by enhancing its pharmacokinetic properties. The sustained release of the synthesized delivery system was evaluated using a kinetic modeling approach along with its biological efficacy. The SEM analysis confirmed the coupling of starch and chitosan and loading of curcumin into the matrix during synthesis. Crystallographic orientation of the delivery system was confirmed by XRD, and the FT-IR data confirmed the successful loading of curcumin. TG analysis indicated that the increment in decomposition temperature of the composite is due to the incorporation of curcumin into the composite. BET analysis revealed the reduction in the BET surface area and pore volume in the synthesized composite, confirming successful loading. The IC50 values of the curcumin and composite for the DPPH assay were 17.81 μg/mL and 38.77 μg/mL, respectively. Composite material has shown enhanced antibacterial activity against Staphylococcus aureus and Pseudomonas aeruginosa at 20 mg/mL, with inhibition zones of 15.56 ± 0.29 and 10.11 ± 0.44 mm, respectively. Allium cepa cytotoxicity assay confirmed that all the synthesized materials promote mitosis at 2.5 mg/mL concentration. The pH and ionic strength responsive release of curcumin was studied by fitting the release data into six kinetic models, including Korsemeyer–Peppas (KP) and Peppas–Sahlin (PS), which has not been reported earlier for the synthesized system. The KP and PS models were selected to interpret the release mechanism based on R2. A combination of Fickian diffusion, relaxation and swelling dominates the curcumin release. According to the KP model, quasi-Fickian diffusion is responsible for the release in acidic and alkaline pH, whereas non-Fickian diffusion occurs at pH 6.7. At higher [NaCl], relaxation is responsible for the release (n > 0.43, kD = 0), while a combination of diffusion and relaxation governs the release at lower [NaCl]. These behaviors are related to the chemical composition of chitosan and the release media, where the electrostatic repulsion, protonation and charge screening contribute to curcumin release.Publication Embargo Joint Resource Allocation and Network Slicing in Hybrid RF/VLC Systems: A Proportional-Fairness Approach(Institute of Electrical and Electronics Engineers Inc., 2026-08-01) Rajahrajasingh, H; Jayakody, D. N. K.; Peha, J. MThis article presents a joint resource allocation and network slicing framework for hybrid radio frequency (RF) and visible light communication (VLC) systems based on a proportional-fairness (PF) objective. The proposed approach formulates a convex optimization problem that jointly allocates bandwidth resources across multiple slices and access technologies to balance aggregate throughput and user fairness. A relaxed PF solution is first derived via projected gradient optimization, followed by two practical rounding strategies: a simple threshold-based method and a greedy assignment heuristic that produce integer allocations with low complexity. Simulation results for a 20 MHz hybrid RF/VLC system (10 MHz/band) demonstrate that the proposed PF-based allocation achieves aggregate sum-rates of approximately 450-500 Mb/s with Jain's fairness indices greater than 0.9, thus outperforming static and rounding baselines by 10%-15% in throughput while maintaining high fairness. The analysis is further extended to asymmetric RF/VLC bandwidth conditions, demonstrating that the proposed PF-based slicing maintains high fairness while scaling throughput effectively. These results confirm that the PF formulation effectively captures the throughput fairness tradeoff and provides near-optimal slicing performance under realistic indoor bandwidth and power constraints.Publication Embargo OcupHI: knowledge-driven colorimetric interpretation framework for high-precision real-time ocular pH diagnostics(Springer Nature, 2026-08-12) Kahandawala, B, S; Sandaruwan, H. H. P. B; Liyanage, P; Dassanayake, R.S; Costha, N.P; Liyanage, R.N; Wijenayake, U; Wijesinghe, R.E; Silva, B.N; Manatunga, D.COcular injuries due to chemical spills pose a substantial concern, representing 10–22% of all ocular trauma. Although precise detection of ocular pH is crucial for determining the optimal medical treatment, many existing methods remain invasive, biased, or insufficiently precise. Reliance on subjective visual assessment of subtle color differences limits the objectivity and hinders high-throughput analysis. Therefore, an advanced colorimetric knowledge-driven ocular pH detection method was developed using a biosensor (OcupHI) based on a Clitoria ternatea (Butterfly Pea) anthocyanin sensing agent. The proposed work delivers fast, high-precision, and easily measurable pH prediction across clinically relevant ranges, while supporting real-time decision support for eye physicians. The pH range from 1 to 12 was tested and compared with six different anthocyanin concentrations: 5, 10, 20, 30, 40, and 50 ppm, and five different machine learning models, namely, Decision Tree (DT), K-Nearest Neighbors (KNN), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Machines (SVM). The results revealed that the 40 ppm anthocyanin concentration trained with the XGBoost model produced the most accurate ocular pH values, achieving superior performance with an overall accuracy of 96%, a significantly higher F1-score for early detection. Experimental validation clearly demonstrates strong predictive accuracy, robustness, and interpretability, highlighting the potential for next-generation ocular diagnostics. Further research findings support Sustainable Development Goal (SDG) 3 – good health and well-being through a real-time ocular pH monitoring kit, and SDG 12 – responsible consumption and production by optimizing the use of the natural colorant anthocyanin for sensor development.Publication Open Access Integrating industry 4.0 and 5.0 technologies in luxury fashion retail interiors: A systematic review of digital transformation, sensory design, and brand storytelling(Elsevier Ltd, 2026-05-12) Ratnayake, Janitha C; Jayasuriya, N; Suraweera, T; De Silva, LEmerging developments in Industry 4.0 and Industry 5.0 are transforming the cultural, sensory, and experiential character of luxury fashion retail interiors. These changes, driven by artificial intelligence, the Internet of Things, augmented and virtual reality, influence how individuals perceive space, form emotional connections, and engage with brand narratives across both physical and digitally enriched environments. Against this backdrop, this review aims to systematically synthesise how Industry 4.0 and Industry 5.0 technologies intersect with sensory experience, spatial design, and narrative expression in luxury fashion retail interiors, addressing the tendency of prior reviews to examine digital technologies or experiential aspects of retail in isolation. The study adopts a systematic literature review approach guided by the SPIDER framework and the PRISMA protocol. Fifty peer reviewed publications published between 2014 and 2025 were analysed to examine how digital transformation, sensory experience, and narrative expression intersect within luxury retail culture. The thematic analysis identified three closely connected areas: branding and customer experience, the interior environment of luxury stores, and the integration of advanced technologies within retail spaces. Together, these themes illustrate how contemporary luxury interiors operate as culturally expressive and emotionally charged settings shaped by new forms of technological mediation. Building on these insights, the study introduces the Multi-Layered Integration Framework, which explains the interaction between digital systems, spatial atmospheres, and brand storytelling in the creation of culturally responsive and human-centred retail interiors. The review contributes to the social sciences and humanities by demonstrating how emerging technologies reshape sensory engagement, symbolic identity, and cultural expression in luxury retail settings, while offering an expanded understanding of human experience within digitally influenced interior environments.Publication Open Access Determinants of under-five mortality in Africa: evidence from a two-decade panel analysis for public health policy(BioMed Central Ltd, 2026-06-17) Rathnasekara, H; Jayathilaka, RBackground: Child survival is a critical indicator for a nation’s health and its progress is important in attaining the Sustainable Development Goals. Understanding the regional and country-specific dynamic and interplay of various determinants of under-five child mortality is vital for the African continent, which remains one of the most vulnerable regions for child mortality globally. Methods: This study investigates the association of economic, health-related, social and demographic, environmental, and infrastructure-related factors with under-five child mortality. It integrates generalisable regional associations using a standard fixed-effects panel model and examines illustrative country-specific associations of the selected determinants through multiple linear regression, based on a balanced panel dataset of 45 countries over a 22-year period. Results: Fixed-effect analysis reveals that the diphtheria-tetanus-pertussis (DTP) immunisation and total fertility rate (TFR) are robust regional determinants of under-five mortality across specifications. While health expenditure, sanitation services, and malaria incidence show significant associations in the baseline model, these findings are sensitive to the inclusion of year fixed effects, suggesting they are influenced by broader temporal trends or common regional shocks rather than serving as stable independent factors within the study period. Conclusion: Regional analysis, which controls for unobserved country-specific heterogeneity over an extended period and is complemented by country-specific analysis, facilitates the formulation of policy implications at both national and international levels. Recommended policy measures include increasing immunisation coverage, implementing malaria control programmes, strengthening community health infrastructure, enhancing girls’ education, promoting widespread access to modern family planning, and improving sanitation services. These strategies are expected to contribute to the progress toward Sustainable Development Goal 3.2 in the African region.