TY - GEN
T1 - Hyperspectral Super-Resolution by Unsupervised Convolutional Neural Network and Sure
AU - Nguyen, Han V.
AU - Ulfarsson, Magnus O.
AU - Sveinsson, Johannes R.
AU - Mura, Mauro Dalla
N1 - Funding Information: This work was supported in part by the University of Iceland Doctoral Fund under Grant 1547-15430, and the Icelandic Research Fund under Grant 207233-051. Publisher Copyright: © 2022 IEEE.
PY - 2022/7/17
Y1 - 2022/7/17
N2 - Recent advances in deep learning (DL) reveal that the structure of a convolutional neural network (CNN) is a good image prior (called deep image prior (DIP)), bridging the model-based and DL-based methods in image restoration. However, optimizing a DIP-based CNN is prone to over-fitting leading to a poorly reconstructed image. This paper derives a loss function based on Stein's unbiased risk estimate (SURE) for unsupervised training of a DIP-based CNN applied to the hyperspectral image (HSI) super-resolution. The SURE loss function is an unbiased estimate of the mean-square-error (MSE) between the clean low-resolution image and the low-resolution estimated image, which relies only on the observed low-resolution image. Experimental results on HSI show that the proposed method not only improves the performance, but also avoids overfitting. Codes are available at https://github.com/hvn2/SURE-MS-HS
AB - Recent advances in deep learning (DL) reveal that the structure of a convolutional neural network (CNN) is a good image prior (called deep image prior (DIP)), bridging the model-based and DL-based methods in image restoration. However, optimizing a DIP-based CNN is prone to over-fitting leading to a poorly reconstructed image. This paper derives a loss function based on Stein's unbiased risk estimate (SURE) for unsupervised training of a DIP-based CNN applied to the hyperspectral image (HSI) super-resolution. The SURE loss function is an unbiased estimate of the mean-square-error (MSE) between the clean low-resolution image and the low-resolution estimated image, which relies only on the observed low-resolution image. Experimental results on HSI show that the proposed method not only improves the performance, but also avoids overfitting. Codes are available at https://github.com/hvn2/SURE-MS-HS
KW - Hyperspectral image
KW - Stein's unbiased risk estimate (SURE)
KW - image fusion
KW - unsupervised CNN
UR - https://www.scopus.com/pages/publications/85140399364
U2 - 10.1109/IGARSS46834.2022.9883576
DO - 10.1109/IGARSS46834.2022.9883576
M3 - Conference contribution
T3 - International Geoscience and Remote Sensing Symposium (IGARSS)
SP - 903
EP - 906
BT - IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022
Y2 - 17 July 2022 through 22 July 2022
ER -