Multidisciplinary AI Research Centre University of Peradeniya
Contact us

Biography

Combines theory-first analysis and practical implementation to develop robust algorithms and low-cost prototypes for remote sensing and energy applications. His work spans hyperspectral unmixing, change detection, and NILM, with an emphasis on modular, edge-deployable solutions that bridge research and real-world deployment.

Associated projects 2

Publications 5

  1. Mamba-FCS: Joint Spatio-Frequency Feature Fusion, Change-Guided Attention, and SeK Inspired Loss for Enhanced Semantic Change Detection in Remote Sensing (opens in a new tab)

    Buddhi Wijenayake, Athulya Ratnayake, Praveen Sumanasekara, Roshan Godaliyadda, Parakrama Ekanayake, Vijitha Herath, Nichula Wasalathilaka

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · 2026

    Journal Accepted Hyperspectral Unmixing DOI 10.1109/JSTARS.2026.3663066

  2. Preprocessing Algorithm Leveraging Geometric Modeling for Scale Correction in Hyperspectral Images for Improved Unmixing Performance (opens in a new tab)

    Praveen Sumanasekara, Athulya Ratnayake, Buddhi Wijenayake, Keshawa Ratnayake, Roshan Godaliyadda, Parakrama Ekanayake, Vijitha Herath

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · 2026

    Journal Accepted Hyperspectral Unmixing DOI 10.1109/JSTARS.2026.3687834

  3. Hyperspectral unmixing with spatial context and endmember ensemble learning with attention mechanism (opens in a new tab)

    RMKL Ratnayake, DMUP Sumanasekara, HMKD Wickramathilaka, GMRI Godaliyadda, HMVR Herath, MPB Ekanayake

    ISPRS Open Journal of Photogrammetry and Remote Sensing · 2025

    Journal Accepted Hyperspectral Unmixing DOI 10.1016/j.ophoto.2025.100086

  4. Precision Spatio-Temporal Feature Fusion for Robust Remote Sensing Change Detection (opens in a new tab)

    Buddhi Wijenayake, Athulya Ratnayake, Praveen Sumanasekara, Nichula Wasalathilaka, Mathivathanan Piratheepan, Roshan Godaliyadda, Mervyn Ekanayake, Vijitha Herath

    International Conference on Industrial and Information Systems 2025 · 2025

    Conference Accepted Hyperspectral Unmixing DOI 10.48550/arXiv.2507.11523

  5. Comparison of Appliance Signature Classification Methods for Non-Intrusive Load Monitoring (opens in a new tab)

    Keshawa Ratnayake, Mahela Pandukabhaya, Praveen Sumanasekara, Uthpala Ratnayake, Roshan Godaliyadda, Janaka Ekanayake, Parakrama Ekanayake, Vijitha Herath

    2024 4th International Conference on Electrical Engineering (EECon) · 2024

    Conference Published AI-Powered Non-Intrusive Load Monitoring (NILM) Implicit Neural Representation DOI 10.1109/EECon64470.2024.10841878