Imran Ashraf

Research & Publications

My work centres on explainable AI: models whose decisions a person can check. I also work on multi-agent reinforcement learning for autonomous driving and on computer vision for health and agriculture.

12 papers4 first-author papersORCIDWeb of Science

2026

  1. Scalable and interpretable text intelligence for fake news detection

    SEET 2026

    First author

  2. Explainable multimodal plant disease classification with prompt-guided transformers

    IEEE DataSciMI 2026

    Second author, presenter

  3. Intelligent simulation and modeling for autonomous vehicles: a review

    IEEE Open Journal of Intelligent Transportation Systems · accepted

    Co-author

  4. AI-powered surveillance systems: a comprehensive review

    WIREs Data Mining and Knowledge Discovery

    Co-author

  5. SA-NET: self-attention for early skin cancer diagnosis

    SEET 2026

    First author

2025

  1. Explainable phishing website detection for secure and sustainable cyber infrastructure

    Nature Scientific Reports

    Corresponding author

  2. Multi-agent reinforcement learning for safe autonomous vehicle platooning

    Springer LNEE, ITSS-IoE 2025 (UK)

    First author, presenter

  3. Wheat disease detection with a shifted window transformer

    IEEE Transactions on Consumer Electronics

    Co-author

  4. CADSNet: building damage assessment from disaster satellite imagery

    IEEE ICCA 2025

    Co-author

  5. Deepfake video detection with Xception-CBAM-BiLSTM

    ICETCS 2025 · accepted

    Co-author

2024

  1. Micro-expression recognition with a Broadbent attention mechanism

    IEEE HONET 2024

    First author

  2. Diabetic foot ulcer detection with SSD and YOLO

    IEEE HONET 2024

    Co-author