Hengwei Zhao's Homepage


Research Interests and Selected Publications

(* Indicates the equal contribution, # Indicates the corresponding author)
(More papers can be found in the Google Scholar)


Weakly Supervised and Trustworthy Learning Theory

  • Hengwei Zhao, Xinyu Wang#, Jingtao Li and Yanfei Zhong.
    Class Prior-Free Positive-Unlabeled Learning with Taylor Variational Loss for Hyperspectral Remote Sensing Imagery.
    Proceedings of the International Conference on Computer Vision (ICCV), 2023. [arXiv] [Paper] [Code]
    (Class prior-free PU Learning with limited training data.)

  • Hengwei Zhao, Yanfei Zhong#, Xinyu Wang and Hong Shu.
    One-Class Risk Estimation for One-Class Hyperspectral Image Classification.
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2023. [Paper] [Code]
    (Distribution imbalanced & distribution overlapped PU Learning.)




Intelligent Remote Sensing Data Interpretation

  • Ziying Liu, Hengwei Zhao, Xinyu Wang, Shaoyu Wang, Jingtao Li and Yanfei Zhong#.
    PU-KBS: A Robust Positive and Unlabeled Learning Framework with Key Band Selection for One-Class Hyperspectral Image Classification.
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024. [Paper]
    (Class prior-free PU learning with key band selection.)

  • Jingtao Li, Xinyu Wang#, Shaoyu Wang, Hengwei Zhao, Yanfei Zhong.
    One Step Detection Paradigm for Hyperspectral Anomaly Detection via Spectral Deviation Relationship Learning.
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024. [Paper]

  • Ge Tang, Xinyu Wang#, Hengwei Zhao, Xin Hu, Guang Jin, Yanfei Zhong.
    Attention in Attention for Hyperspectral with High Spatial Resolution (H2) Image Classification.
    IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024. [Paper]

  • Jingtao Li, Xinyu Wang#, Hengwei Zhao and Yanfei Zhong.
    Anomaly Segmentation for High-Resolution Remote Sensing Images Based on Pixel Descriptors.
    Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023. [Paper]

  • Xin He, Hengwei Zhao, Xinyu Wang and Yanfei Zhong#.
    A Rapid One-Stage End to End Hyperspectral Target Detection Model.
    Proceedings of 2022 10th China Conference on Command and Control (C2 CHINA), 2022. [Paper]
    (A PU learning-based rapid one-stage end to end hyperspectral target detection model.)




Ecological Environmental and Natural Hazard Monitoring

  • Hengwei Zhao*, Junjue Wang*, Yang Pan, Ailong Ma, Xinyu Wang and Yanfei Zhong#.
    Progressive Label Refinement-Based Distribution Adaptation Framework for Landslide Detection.
    Proceedings of the Second Workshop on Complex Data Challenges in Earth Observation co-located with 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence (IJCAI-ECAI, CDCEO), 2022. [Paper]
    (The technical report for the IJCAI CDCEO Landslide4Sense Competiton.)

  • Jingtao Li, Xinyu Wang#, Hengwei Zhao, Xin Hu and Yanfei Zhong.
    Detecting Pine Wilt Disease at the Pixel Level from High Spatial and Spectral Resolution UAV-borne Imagery in Complex Forest Landscapes using Deep One-class Classification.
    International Journal of Applied Earth Observation and Geoinformation (JAG), 2022. [Paper]
    (Research on the application of PU learning technology in pest and disease detection from UAV-borne hyperspectral images.)

  • Hengwei Zhao, Yanfei Zhong#, Xinyu Wang, Xin Hu, Chang Luo, Mark Boitt, Rami Piiroinen, Liangpei Zhang, Janne Heiskanen and Petri Pellikka.
    Mapping the Distribution of Invasive Tree Species using Deep One-class Classification in the Tropical Montane Landscape of Kenya.
    ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS P&RS), 2022. [Paper] [Code]
    (Research on the application of PU Learning technology in invasive tree species detection from airborne hyperspectral images.)

  • Lei Lei, Xinyu Wang#, Yanfei Zhong, Hengwei Zhao, Xin Hu and Chang Luo.
    DOCC: Deep One-class Crop Classification via Positive and Unlabeled Learning for Multi-modal Satellite Imagery.
    International Journal of Applied Earth Observation and Geoinformation (JAG), 2021. [Paper]
    (Research on the application of PU learning technology in crop extraction from spaceborne hyperspectral images.)