< Home       

Resolution Enhancement

Introduction

This research project is mainly focusing on various applications of image resolution enhancement, including natural image super-resolution, the fusion of panchromatic and multispectral images from satellites (also called pansharpening), and the fusion of multispectral and hyperspectral images (also called MSI&HSI fusion or HSI super-resolution), etc. We previously utilized the tools of PDE and optimization modeling to conduct the above tasks. Recently, we partially turn the direction to formulate the novel and practical deep-learning approaches for the job of resolution enhancement, as well as obtain state-of-the-art outcomes. Until now, our group has published many related papers on the flagship journals and conferences, e.g., IEEE TIP, Infor. Fusion, IEEE TCSVT, IEEE ICIP, IEEE IGARSS, etc.

 
Pansharpening + MSI&HSI Fusion:
  • BAM: Bilateral Activation Mechanism for Image Fusion
    Zi-Rong Jin, Liang-Jian Deng*, Tian-Jing Zhang, Xiaoxu Jin
    ACM International Conference on Multimedia (ACM MM) , 2021.   [pdf]  [Project Page]
  • Hyperspectral Image Super-resolution via Deep Spatio-spectral Convolutional Neural Networks
    Jin-Fan Hu, Ting-Zhu Huang*, Liang-Jian Deng*, Tai-Xiang Jiang, Gemine Vivone, Jocelyn Chanussot
    IEEE Transactions on Neural Networks and Learning Systems , 2021.   [pdf]  [Project Page]
  • Detail Injection-based Deep Convolutional Neural Networks for Pansharpening
    Liang-Jian Deng, Gemine Vivone*, Cheng Jin, Jocelyn Chanussot
    IEEE Transactions on Geoscience and Remote Sensing, 2020.    [pdf]  [Project Page]
  • A New Variational Approach Based on Proximal Deep Injection and Gradient Intensity Similarity for Spatio-spectral Image Fusion
    Zhong-Chen Wu, Ting-Zhu Huang*, Liang-Jian Deng*, Gemine Vivone, Jia-Qing Miao, Jin-Fan Hu, Xi-Le Zhao
    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020.    [pdf]  [Project Page]
  • Hyperspectral Image Superresoluition Using Unidirectional Total Variation With Tucker Decomposition
    Ting-Xu, Ting-Zhu Huang*, Liang-Jian Deng*, Xi-Le Zhao, Jie Huang
    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020.    [pdf]  [Matlab_Code]
  • A variational pansharpening approach based on reproducible kernel Hilbert space and Heaviside function
    Liang-Jian Deng, Gemine Vivone, Weihong Guo, Mauro Dalla Mura, Jocelyn Chanussot
    IEEE Transactions on Image Processing, 27(9): 4330-4344, 2018.   [pdf]  [Project Page]
  • The fusion of panchromatic and multispectral remote sensing images via tensor-based sparse modeling and hyper-Laplacian prior
    Liang-Jian Deng, Minyu Feng, Xue-Cheng Tai
    Information Fusion, 52: 76-89, 2019.   [pdf]  [Matlab_Code]
  • A Rudin-Osher-Fatemi Model-Based Pansharpening Approach Using RKHS and AHF Representation
    J Liu, Liang-Jian Deng, F Fang, T Zeng
    East Asian Journal on Applied Mathematics, 2019.   [pdf]
  • A Framelet-Based Iterative Pan-Sharpening Approach
    Zi-Yao Zhang, Ting-Zhu Huang, Liang-Jian Deng, Jie Huang, Xi-Le Zhao, Chao-Chao Zheng
    Remote Sensing, 10(4): 622, 2018.   [pdf]
  • Image Fusion via Dynamic Gradient Sparsity and Anisotropic Spectral-Spatial Total Variation
    Chao-Chao Zheng, Ting-Zhu Huang, Liang-Jian Deng, Xi-Le, Zhao, Hong-Xia Dou
    IEEE International Conference on Image Processing (ICIP), 2017.   [pdf]
  • Pan-sharpening via ROG-Based Filtering
    Zi-Yao Zhang, Ting-Zhu Huang, Liang-Jian Deng, Jie Huang, Hong-Xia Dou
    IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2019.
 
Natural Image Super-Resolution (SR):
  • Single image super-resolution via an iterative reproducing kernel Hilbert space method
    Liang-Jian Deng, Weihong Guo, Ting-Zhu Huang
    IEEE Transactions on Circuits and Systems for Video Technology, 26(11): 2001-2014, 2016.   [pdf]  [Matlab_Code]
  • Single image super-resolution by approximate Heaviside functions
    Liang-Jian Deng, Weihong Guo, Ting-Zhu Huang
    Information Sciences, 348:107-123, 2016.   [pdf]
  • Single image super-resolution based on approximated Heaviside functions and iterative refinement
    Xiyu Wang, Ting-Zhu Huang, Liang-Jian Deng
    PLoS One, 2018
  • Super-resolution via a fast deconvolution with kernel estimation
    Han Yu, Ting-Zhu Huang, Liang-Jian Deng
    EURASIP Journal on Image and Video Processing, 2017(3): 1-11, 2016