Research Papers - Department of Electrical and Electronic Engineering
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/679
Browse
134 results
Filters
Advanced Search
Filter by
Settings
Search Results
Publication Embargo Autoencoder based data clustering for identifying anomalous repetitive hand movements, and behavioral transition patterns in children(Springer Science and Business Media Deutschland GmbH, 2025-01-21) Wedasingha, N; Samarasinghe, P; Senevirathna, L; Papandrea, M; Puiatti, AThe analysis of repetitive hand movements and behavioral transition patterns holds particular significance in detecting atypical behaviors in early child development. Early recognition of these behaviors holds immense promise for timely interventions, which can profoundly impact a child’s well-being and future prospects. However, the scarcity of specialized medical professionals and limited facilities has made detecting these behaviors and unique patterns challenging using traditional manual methods. This highlights the necessity for automated tools to identify anomalous repetitive hand movements and behavioral transition patterns in children. Our study aimed to develop an automated model for the early identification of anomalous repetitive hand movements and the detection of unique behavioral patterns. Utilizing autoencoders, self-similarity matrices, and unsupervised clustering algorithms, we analyzed skeleton and image-based features, repetition count, and frequency of repetitive child hand movements. This approach aimed to distinguish between typical and atypical repetitive hand movements of varying speeds, addressing data limitations through dimension reduction. Additionally, we aimed to categorize behaviors into clusters beyond binary classification. Through experimentation on three datasets (Hand Movements in Wild, Updated Self-Stimulatory Behaviours, Autism Spectrum Disorder), our model effectively differentiated between typical and atypical hand movements, providing insights into behavioral transitional patterns. This aids the medical community in understanding the evolving behaviors in children. In conclusion, our research addresses the need for early detection of atypical behaviors through an automated model capable of discerning repetitive hand movement patterns. This innovation contributes to early intervention strategies for neurological conditionsPublication Open Access Robust Distributed Collaborative Beamforming for WSANs in Dual-Hop Scattered Environments with Nominally Rectangular Layouts(Multidisciplinary Digital Publishing Institute (MDPI), 2025-03-19) Ben Smida, O; Affes, S; Jayakody, D; Nizam, YWe introduce a robust distributed collaborative beamforming (RDCB) approach for addressing channel estimation challenges in dual-hop transmissions within wireless sensor and actuator networks (WSANs) of K nodes. WSANs enhance wireless communication by reducing data transmission, latency, and energy consumption while optimizing network load through integrated sensing and actuation. The source S transmits signals to the WSAN, where nodes relay them to the destination D using beamforming weights to minimize noise and preserve signal integrity. These weights depend on channel state information (CSI), where estimation errors degrade performance. We develop RDCB solutions for three first-hop propagation scenarios—monochromatic [line-of-sight (LoS)] or “M”, bichromatic (moderately scattered) or “B”, and polychromatic (highly scattered) or “P”—while assuming a monochromatic LoS or “M” link for the second hop between the nodes and the far-field destination. Termed MM-RDCB, BM-RDCB, and PM-RDCB, respectively (“X” and “Y” in XY-RDCB—for X (Formula presented.) and Y (Formula presented.) —refer to the chromatic natures of the first- and second-hop channels, respectively, to which a specific RDCB solution is tailored), these solutions leverage asymptotic approximations for large K values and the nodes’ geometric symmetries. Our distributed solutions allow local weight computation, enhancing spectral and power efficiency. Simulation results show significant improvements in the signal-to-noise ratio (SNR) and robustness versus WSAN node placement errors, making the solutions well suited for emerging 5G and future 5G+/6G and Internet of Things (IoT) applications for different challenging environments.Publication Open Access Efficient Hotspot Detection in Solar Panels via Computer Vision and Machine Learning(Multidisciplinary Digital Publishing Institute (MDPI), 2025-07-15) Fernando, N; Seneviratne, L; Weerasinghe, N; Rathnayake, N; Hoshino, YSolar power generation is rapidly emerging within renewable energy due to its cost-effectiveness and ease of deployment. However, improper inspection and maintenance lead to significant damage from unnoticed solar hotspots. Even with inspections, factors like shadows, dust, and shading cause localized heat, mimicking hotspot behavior. This study emphasizes interpretability and efficiency, identifying key predictive features through feature-level and What-if Analysis. It evaluates model training and inference times to assess effectiveness in resource-limited environments, aiming to balance accuracy, generalization, and efficiency. Using Unmanned Aerial Vehicle (UAV)-acquired thermal images from five datasets, the study compares five Machine Learning (ML) models and five Deep Learning (DL) models. Explainable AI (XAI) techniques guide the analysis, with a particular focus on MPEG (Moving Picture Experts Group)-7 features for hotspot discrimination, supported by statistical validation. Medium Gaussian SVM achieved the best trade-off, with 99.3% accuracy and 18 s inference time. Feature analysis revealed blue chrominance as a strong early indicator of hotspot detection. Statistical validation across datasets confirmed the discriminative strength of MPEG-7 features. This study revisits the assumption that DL models are inherently superior, presenting an interpretable alternative for hotspot detection; highlighting the potential impact of domain mismatch. Model-level insight shows that both absolute and relative temperature variations are important in solar panel inspections. The relative decrease in “blueness” provides a crucial early indication of faults, especially in low-contrast thermal images where distinguishing normal warm areas from actual hotspot is difficult. Feature-level insight highlights how subtle changes in color composition, particularly reductions in blue components, serve as early indicators of developing anomalies.Publication Open Access A Context-Aware Doorway Alignment and Depth Estimation Algorithm for Assistive Wheelchairs(Multidisciplinary Digital Publishing Institute (MDPI), 2025-07-17) Tennekoon, S; Wedasingha, N; Welhenge, A; Abhayasinghe, N; Murray, INavigating through doorways remains a daily challenge for wheelchair users, often leading to frustration, collisions, or dependence on assistance. These challenges highlight a pressing need for intelligent doorway detection algorithm for assistive wheelchairs that go beyond traditional object detection. This study presents the algorithmic development of a lightweight, vision-based doorway detection and alignment module with contextual awareness. It integrates channel and spatial attention, semantic feature fusion, unsupervised depth estimation, and doorway alignment that offers real-time navigational guidance to the wheelchairs control system. The model achieved a mean average precision of 95.8% and a F1 score of 93%, while maintaining low computational demands suitable for future deployment on embedded systems. By eliminating the need for depth sensors and enabling contextual awareness, this study offers a robust solution to improve indoor mobility and deliver actionable feedback to support safe and independent doorway traversal for wheelchair users.Publication Open Access Electromagnetic Continuously Variable Transmission (EMCVT) System for Precision Torque Control in Human-Centered Robotic Applications(Multidisciplinary Digital Publishing Institute (MDPI), 2025-09-08) Madusankha, I; Jayaweera, P. N; Kahatapitiya, N. S; Sampath, P; Weeraratne, A; Subasinghage, K; Liyanage, C; Wijethunge, A; Ravichandran, N. K; Wijesinghe, R. EIn human-centered robotic applications, safety, efficiency, and adaptability are critical for enabling effective interaction and performance. Incorporating electromagnetic continuously variable transmission (EM-CVT) systems into robotic designs enhances both safety and precise, adaptable motion control. The flexible power transmission offered by CVTs allows robots to operate across diverse environments, supporting various tasks, human interaction, and safe collaboration. This study presents a CVT-based mechanical subsystem developed using two cones and an intermediate belt-driven transmission mechanism, providing efficient power and motion transfer. The control subsystem consists of six strategically positioned electromagnets energized by signals from a microcontroller. This electromagnetic actuation enables rapid and precise adjustments to the transmission ratio, enhancing overall system performance. A linear relationship between slip percentage and gear ratio was observed, indicating that the control system achieves stable and efficient operation, with a measured power consumption of 2.95 W per electromagnet. Future work will focus on validating slip performance under dynamic loading conditions, integrating the system into robotic platforms, and optimizing materials and control strategies to enable broader real-world deployment.Publication Embargo Environmental forensics of the X-press pearl disaster: Uncovering the internal micro-structural transformations in marine microplastics(Elsevier B.V., 2025-07-15) Jayasekara, P.M; Abhishek, P; Kahatapitiya, N. S; Weerasinghe, M; Kahandawala, B. S; Silva, B. N; Wijenayake, U; Rajapaksha, A.U; Wijesinghe, R. E; Vithanage, MThe MV X-Press Pearl (XPP) maritime disaster on May 25, 2021, released approximately 75 billion microplastic (MP) nurdles into the Indian Ocean and degraded due to the elevated temperatures, a cocktail of chemicals, physical abrasions, and environmental factors. While degradation-induced surface-level chemical and morphological changes were well documented, internal degradation remains largely unexplored. This study highlights the utilization of high-resolution optical coherence tomography (OCT) as a purely non-destructive imaging modality to discover profound internal alterations in the micrometer range, such as internal hollow regions, cracks, and voids in MP nurdles subjected to different degrees of degradation. The dark pixel intensity probability density corresponds to the degraded areas, increased from 0.0019 (pristine nurdle) to 0.0135–0.5252 for thermal degradation, 0.0878–0.3134 for chemical degradation, and 0.1291–0.2179 for mechanical degradation, indicating progressive internal degradation. Attenuated total reflectance fourier transform infrared (ATR-FTIR) spectroscopy analysis confirmed that all the nurdles are polyethylene (PE) and revealed that extreme conditions lead to the formation of new functional groups, including hydroxyl bands and carbonyl bands, even though PE is highly resistant to degradation. The integration of high-resolution OCT imaging with FTIR analysis provides novel insights into the interconnection between micrometer-scale internal physical alterations and associated chemical modifications of MP nurdles resulting from environmental degradation. These findings highlight the potential of this OCT-FTIR integrated approach for advancing the understanding of MP degradation and its long-term environmental impacts.Publication Open Access Facial identity recognition using StyleGAN3 inversion and improved tiny YOLOv7 model(Nature Research, 2025-03-17) Kumar, A; Bhattacharjee, S; Kumar, Ambrish; Jayakody, D. N.KFacial identity recognition is one of the challenging problems in the domain of computer vision. Facial identity comprises the facial attributes of a person’s face ranging from age progression, gender, hairstyle, etc. Manipulating facial attributes such as changing the gender, hairstyle, expressions, and makeup changes the entire facial identity of a person which is often used by law offenders to commit crimes. Leveraging the deep learning-based approaches, this work proposes a one-step solution for facial attribute manipulation and detection leading to facial identity recognition in few-shot and traditional scenarios. As a first step towards performing facial identity recognition, we created the Facial Attribute Manipulation Detection (FAM) Dataset which consists of twenty unique identities with thirty-eight facial attributes generated by the StyleGAN3 inversion. The Facial Attribute Detection (FAM) Dataset has 11,560 images richly annotated in YOLO format. To perform facial attribute and identity detection, we developed the Spatial Transformer Block (STB) and Squeeze-Excite Spatial Pyramid Pooling (SE-SPP)-based Tiny YOLOv7 model and proposed as FIR-Tiny YOLOv7 (Facial Identity Recognition-Tiny YOLOv7) model. The proposed model is an improvised variant of the Tiny YOLOv7 model. For facial identity recognition, the proposed model achieved 10.0% higher mAP in the one-shot scenario, 30.4% higher mAP in the three-shot scenario, 15.3% higher mAP in the five-shot scenario, and 0.1% higher mAP in the traditional 70% − 30% split scenario as compared to the Tiny YOLOv7 model. The results obtained with the proposed model are promising for general facial identity recognition under varying facial attribute manipulation.Publication Embargo Brewing plastics: OCT reveals microplastic release from nylon tea bags in simulated brewed tea infusions(Royal Society of Chemistry, 2026-02-12) Jayasekara, P.M; Abhishek, P; Kahandawala,B.S; Damith, N; Weerasinghe, M; Kahatapitiya, N.S; Silva, B.N; Karunaratne, S; Wijesinghe, R.E; Wijenayake, UThe release of microplastics (MPs) from nylon tea bags poses a critical concern for human exposure; however,their detection and quantification remain challenging especially in beverage matrices, and hence, this study pioneers the use of high-resolution optical coherence tomography (OCT) integrated with an image processing algorithm to rapidly detect and quantify the size and count of the MPs directly in the water extractions simulating tea brewing. The water extractions prepared by simulating tea brewing conditions, hot (100 °C, 1–5min), cold (2 °C, 1 h), and ambient (30 °C, 1 h), were observed employing OCT imaging and validated through Nile Red (NR) staining and digital microscopy. The nylon tea bags steeped in hot water for 5 minutes released 16 000 to 24 000 LMPs (>30 mm) and SMPs (12–30 mm) per millilitre. The estimated daily intake (EDI) of MPs indicates a higher exposure for children (ranging from 0.201 to 0.349 mm3 kg−1 day−1 ) compared to adults (0.046 to 0.080 mm3 kg−1 day−1 ). In contrast, cold brewing for 1 hour released fewer LMPs but an equal quantity of small MPs (SMPs) compared to hot brewing. This OCT-based approach offers a rapid, versatile platform for the detection and quantification of MPs from diverse packaging materials and provides a powerful tool for comprehensive risk assessment when combined with chemical and toxicological analyses.Publication Embargo Tamper Proof Smart Energy Metering System with Power Factor Correction and Surge Protection(IEEE, 2022-10-28) Sandarenu, E.A.P.; Jayasekara, S; Jathunga, T. GCurrently, Sri Lankan electricity suppliers have adapted a postpaid tariff system, in which the consumers with a contract demand equal to or less than 42 kVA are only charged for the active power consumption. This tariff method has several disadvantages for both consumers and electrical suppliers in many ways. Lower power factors, Electricity theft, tampered meters, avoided bill payments, and manual meter reading procedures have caused considerable losses for the electricity suppliers. Similarly, consumers are also discomforted by the postpaid nature of this tariff method since it has caused long queues in the bill payment locations. To overcome above problems, this paper presents a smart energy meter with automated billing capabilities, power factor correction, and tamper protection. This meter is currently intended only for single-phase consumers, in which consumers are allowed to choose the payment method from prepaid or postpaid according to their preference.Publication Embargo Performance Comparison of Sea Fish Species Classification using Hybrid and Supervised Machine Learning Algorithms(IEEE, 2022-10-04) Nalmi, R; Rathnayake, N; Mampitiya, L.IIn the domain of autonomous underwater vehicles, the classification of objects underwater is critical. The hazy effect of the medium causes this obstacle, and these effects are directed by the dissolved particles that lead to the reflecting and scattering of light during the formation process of the image. This paper mainly focuses on exploring the best possible image classifier for the underwater images of the different fish species. The classifications were carried out by different hybrid and supervised machine learning algorithms such as Support Vector Machine (SVM), Random Forest (RF), Neural Networks (NN), Logistic Regression (LR), Decision Tree (DT), and Naive Bayes (NB). This study compares the algorithms’ accuracy and time and analyzes crucial features to decide the most optimal algorithm. Furthermore, the results of this paper depict that using dimension reduction methods such as PCA and LDA increases the accuracy of some algorithms. Random Forest was able to outperforms with a higher accuracy of 99.89% with the proposed feature extraction methods.