Low Photon Budget Phase Retrieval with Perceptual Loss Trained Deep Neural Networks
Preprint
Abstract
Deep neural networks (DNNs) are efficient solvers for ill-posed problems and have been shown to outperform classical optimization techniques in several computational imaging problems. DNNs are trained by solving an optimization problem implies the choice of an appropriate loss function, i.e., the function to optimize. In a recent paper [A. Goy \textit{et al.}, Phys. Rev. Lett. 121(24), 243902 (2018)], we showed that DNNs trained with the negative Pearson correlation coefficient as the loss function are particularly fit for photon-starved phase retrieval problems, which are fundamentally affected by strong Poison noise. In this paper we demonstrate that the use of a perceptual loss function significantly improves the reconstructions.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Low Photon Count Phase Retrieval Using Deep LearningPhysical Review Letters, 2018
- CNN-Based Projected Gradient Descent for Consistent CT Image ReconstructionIEEE Transactions on Medical Imaging, 2018
- Computational optical tomography using 3-D deep convolutional neural networksOptical Engineering, 2018
- Propagation based phase retrieval of simulated intensity measurements using artificial neural networksJournal of Optics, 2018
- Lensless computational imaging through deep learningOptica, 2017
- Deep Convolutional Neural Network for Inverse Problems in ImagingIEEE Transactions on Image Processing, 2017
- Optical Tomographic Image Reconstruction Based on Beam Propagation and Sparse RegularizationIEEE Transactions on Computational Imaging, 2016
- Image Super-Resolution Using Deep Convolutional NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Learning approach to optical tomographyOptica, 2015
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004