Imaging Science Ph.D. Defense: Sihan Huang
![imaging science ph.d. defense sihan huang](/events/sites/rit.edu.events/files/styles/new_event_image/public/eventimages/CIS%20Ph.D%20Defense%20RIT%20Event%20Page%20Hero%20Banner_13.jpg?itok=CI0QT1BN)
Ph.D. Dissertation Defense
Radiometrically-Accurate Hyperspectral Data Sharpening
Sihan Huang
Imaging Science Ph.D. Candidate
Chester F. Carlson Center for Imaging Science, RIT
Advisor: Dr. David Messinger
Register Here for Zoom Link
Abstract:
Improving the spatial resolution of hyperpsectral image (HSI) has traditionally been an important topic in the field of remote sensing. Many approaches have been proposed based on various theories including component substitution, multiresolution analysis, spectral unmixing, Bayesian probability, and tensor representation. However, these methods have some common disadvantages, such as that they are not robust to different up-scale ratio and they have little concern for the per-pixel radiometric accuracy of the sharpened image. Moreover, many learning-based methods have been proposed through decades of innovations, but most of them require a large set of training pairs, which is unpractical for many real tasks. To solve these problems, we firstly proposed an unsupervised Laplacian Pyramid Fusion Network (LPFNet) to generate a radiometrically-accurate high-resolution hyperspectral image (HR-HSI). First, with the low-resolution hyperspectral image (LR-HSI) and the high-resolution multispectral image (HR-MSI), the preliminary HR-HSI is calculated via linear regression. Next, the high-frequency details of the preliminary HR-HSI are estimated via the subtraction between it and the CNN-generated-blurry version. By injecting the details to the output of the generative CNN with the low-resolution hyperspectral image (LR-HSI) as input, the final HR-HSI is obtained. Furthermore, an unsupervised cascade fusion network (UCFNet) is designed to sharpen the LR-HSI covers the Vis-NIR-SWIR bands. First, the preliminary high-resolution VNIR hyperspectral image (HR-VNIR-HSI) is obtained with a conventional hyperspectral algorithm. Then, the HR-MSI, the preliminary HR-VNIR-HSI, and the LR-SWIR-HSI are passed to the generative convolutional neural network to produce a HR-HSI. Experiments are conducted on both LPFNet and UCFNet with different datasets, up-scale ratios and state-of-the-art baselines,results demonstrate that proposed methods outperform the competitors among all cases in terms of spectral and spatial accuracy.
Intended Audience:
Undergraduates, graduates, and experts. Those with interest in the topic.
To request an interpreter, please visit https://myaccess.rit.edu
Event Snapshot
When and Where
Who
Open to the Public
Interpreter Requested?
No