Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution
Hyperspectral imaging is a crucial tool in remote sensing, which captures far more spectral information than standard color images. However, the increase in spectral information comes at the cost of spatial resolution. Super-resolution is a popular technique where the goal is to generate a high-reso...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10319126/ |
| _version_ | 1827398995790528512 |
|---|---|
| author | Alexander Ulrichsen Paul Murray Stephen Marshall Moncef Gabbouj Serkan Kiranyaz Mehmet Yamac Nour Aburaed |
| author_facet | Alexander Ulrichsen Paul Murray Stephen Marshall Moncef Gabbouj Serkan Kiranyaz Mehmet Yamac Nour Aburaed |
| author_sort | Alexander Ulrichsen |
| collection | DOAJ |
| description | Hyperspectral imaging is a crucial tool in remote sensing, which captures far more spectral information than standard color images. However, the increase in spectral information comes at the cost of spatial resolution. Super-resolution is a popular technique where the goal is to generate a high-resolution version of a given low-resolution input. The majority of modern super-resolution approaches use convolutional neural networks (CNNs). However, convolution itself is a linear operation and the networks rely on the nonlinear activation functions after each layer to provide the necessary nonlinearity to learn the complex underlying function. This means that CNNs tend to be very deep to achieve the desired results. Recently, self-organized operational neural networks (ONNs) have been proposed that aim to overcome this limitation by replacing the convolutional filters with learnable nonlinear functions through the use of MacLaurin series expansions. This work focuses on extending the convolutional filters of a popular super-resolution model to more powerful operational filters to enhance the model performance on hyperspectral images (HSIs). We also investigate the effects that residual connections and different normalization types have on this type of enhanced network. Despite having fewer parameters than their convolutional network equivalents, our results show that ONNs achieve superior super-resolution performance on small HSI datasets. |
| first_indexed | 2024-03-08T19:37:56Z |
| format | Article |
| id | doaj.art-0f52d6eb2c5d40ac9d8052a8781eeee2 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T19:37:56Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-0f52d6eb2c5d40ac9d8052a8781eeee22023-12-26T00:01:07ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01171470148410.1109/JSTARS.2023.333327410319126Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-ResolutionAlexander Ulrichsen0https://orcid.org/0009-0007-9838-1882Paul Murray1https://orcid.org/0000-0002-6980-9276Stephen Marshall2https://orcid.org/0000-0001-7079-5628Moncef Gabbouj3https://orcid.org/0000-0002-9788-2323Serkan Kiranyaz4https://orcid.org/0000-0003-1551-3397Mehmet Yamac5https://orcid.org/0000-0002-1681-6931Nour Aburaed6https://orcid.org/0000-0002-5906-0249Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K.Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K.Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K.Faculty of Information Technology and Communication Sciences, Tampere University, Tampere, FinlandElectrical Engineering Department, College of Engineering, Qatar University, Doha, QatarFaculty of Information Technology and Communication Sciences, Tampere University, Tampere, FinlandDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow, U.K.Hyperspectral imaging is a crucial tool in remote sensing, which captures far more spectral information than standard color images. However, the increase in spectral information comes at the cost of spatial resolution. Super-resolution is a popular technique where the goal is to generate a high-resolution version of a given low-resolution input. The majority of modern super-resolution approaches use convolutional neural networks (CNNs). However, convolution itself is a linear operation and the networks rely on the nonlinear activation functions after each layer to provide the necessary nonlinearity to learn the complex underlying function. This means that CNNs tend to be very deep to achieve the desired results. Recently, self-organized operational neural networks (ONNs) have been proposed that aim to overcome this limitation by replacing the convolutional filters with learnable nonlinear functions through the use of MacLaurin series expansions. This work focuses on extending the convolutional filters of a popular super-resolution model to more powerful operational filters to enhance the model performance on hyperspectral images (HSIs). We also investigate the effects that residual connections and different normalization types have on this type of enhanced network. Despite having fewer parameters than their convolutional network equivalents, our results show that ONNs achieve superior super-resolution performance on small HSI datasets.https://ieeexplore.ieee.org/document/10319126/Hyperspectral imagingoperational neural networks (ONNs)super-resolution |
| spellingShingle | Alexander Ulrichsen Paul Murray Stephen Marshall Moncef Gabbouj Serkan Kiranyaz Mehmet Yamac Nour Aburaed Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Hyperspectral imaging operational neural networks (ONNs) super-resolution |
| title | Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution |
| title_full | Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution |
| title_fullStr | Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution |
| title_full_unstemmed | Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution |
| title_short | Operational Neural Networks for Parameter-Efficient Hyperspectral Single-Image Super-Resolution |
| title_sort | operational neural networks for parameter efficient hyperspectral single image super resolution |
| topic | Hyperspectral imaging operational neural networks (ONNs) super-resolution |
| url | https://ieeexplore.ieee.org/document/10319126/ |
| work_keys_str_mv | AT alexanderulrichsen operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT paulmurray operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT stephenmarshall operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT moncefgabbouj operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT serkankiranyaz operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT mehmetyamac operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution AT nouraburaed operationalneuralnetworksforparameterefficienthyperspectralsingleimagesuperresolution |