Unrolling-Based Deep Reconstruction for Compressive Spectral Imaging
| Autor: | Figen S. Oktem, Can Deniz Bezek |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Point spread function
medicine.medical_specialty Computer science business.industry Deep learning Multispectral image ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Hyperspectral imaging Inverse problem Reconstruction method Spectral imaging Diffractive lens medicine Computer vision Artificial intelligence business ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | OSA Imaging and Applied Optics Congress 2021 (3D, COSI, DH, ISA, pcAOP). |
| DOI: | 10.1364/cosi.2021.cm2e.3 |
| Popis: | We develop a novel deep learning-based reconstruction method for compressive spectral imaging with diffractive lenses. The method incorporates U-Net based 3D deep prior to the model-based reconstruction through unrolling. Results illustrate the state-of-the-art performance of our method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|