Novel Techniques for Hyperspectral Image Super-resolution in Remote Sensing
Implementing Organization
SVNIT, Surat and https://www.svnit.ac.in/
Principal Investigator
Dr. Anugrah Jain
SVNIT, Surat and https://www.svnit.ac.in/
ajain@coed.svnit.ac.in
CO-Principal Investigator
Mrs. Meenakshi Sarkar
SVNIT, Surat and https://www.svnit.ac.in/
Project Overview
This project aims to develop an efficient super-resolution framework for hyperspectral images (HSI) leveraging deep learning solutions. The proposed framework will utilize autoencoders, transformers, and attention models for spatial super-resolution of input HSI. The additional spatial information will be taken from the respective multispectral images and their transformer-based super-resolution. An attention mechanism with maximum a posterior based modelling will be used to learn spectral parameters for HSI-MSI fusion. In the deliverables, a tool will be provided which will take HSI and MSI images as input and generate HSI with high spatial resolution as output.
Achievements
The project being undertaken by ISRO in collaboration with academia is directly or indirectly beneficial to the society.