Image Restoration Using Total Variation Regularized Deep Image Prior

Liu, Jiaming; Sun, Yu; Xu, Xiaojian; Kamilov, Ulugbek S.

doi:10.1109/icassp.2019.8682856

Cited by 136 publications

(120 citation statements)

References 31 publications

Supporting

Mentioning

107

Contrasting

Order By: Relevance

“…6. Note that the DIP reconstruction contains a small amount of TV regularization, which we found produced better results than DIP alone [50]. While in the x y dimensions (Fig.…”

Section: Biological Simulation Resultsmentioning

confidence: 57%

Diffraction tomography with a deep image prior

Zhou¹,

Horstmeyer²

2020

Opt. Express

View full text Add to dashboard Cite

We present a tomographic imaging technique, termed Deep Prior Diffraction Tomography (DP-DT), to reconstruct the 3D refractive index (RI) of thick biological samples at high resolution from a sequence of low-resolution images collected under angularly varying illumination. DP-DT processes the multi-angle data using a phase retrieval algorithm that is extended by a deep image prior (DIP), which reparameterizes the 3D sample reconstruction with an untrained, deep generative 3D convolutional neural network (CNN). We show that DP-DT effectively addresses the missing cone problem, which otherwise degrades the resolution and quality of standard 3D reconstruction algorithms. As DP-DT does not require pre-captured data or pre-training, it is not biased towards any particular dataset. Hence, it is a general technique that can be applied to a wide variety of 3D samples, including scenarios in which large datasets for supervised training would be infeasible or expensive. We applied DP-DT to obtain 3D RI maps of bead phantoms and complex biological specimens, both in simulation and experiment, and show that DP-DT produces higher-quality results than standard regularization techniques. We further demonstrate the generality of DP-DT, using two different scattering models, the first Born and multi-slice models. Our results point to the potential benefits of DP-DT for other 3D imaging modalities, including X-ray computed tomography, magnetic resonance imaging, and electron microscopy.

show abstract

“…6. Note that the DIP reconstruction contains a small amount of TV regularization, which we found produced better results than DIP alone [50]. While in the x y dimensions (Fig.…”

Section: Biological Simulation Resultsmentioning

confidence: 57%

Diffraction tomography with a deep image prior

Zhou¹,

Horstmeyer²

2020

Opt. Express

View full text Add to dashboard Cite

show abstract

“…Of interest is the question if DIP can be further improved by adjusting the training objective. For instance, regularization of DIP has been discussed in [7] for TV and in [8] for learned regularization. Similarly, the work [9] proposes to combine DIP with the concept of regularization by denoising introduced in [10].…”

Section: Dip: Background and Prior Workmentioning

confidence: 99%

“…We illustrate this by considering TV. It was shown [7] that TV regularization can further improve the performance of DIP. For a 2D image I, a TV regularizer is formulated as…”

Section: Tv Regularizationmentioning

confidence: 99%

Sub-Dip: Optimization On A Subspace With Deep Image Prior Regularization And Application To Superresolution

Sagel¹,

Roumy²,

Guillemot³

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

“…Several recent studies have in fact shown that CNNs can serve as excellent priors for a wide range of imaging problems. [15][16][17][18][19][20] Our results on experimental and synthetic data show that the proposed CNN-based R * 2 computation method not only reduces the computation time by several orders of magnitude compared with the direct voxel-wise computation, but also significantly improves the computation outcome quality, thanks to the powerful regularization ability of our trained deep CNNs.…”

Section: Introductionmentioning

confidence: 99%

Deep learning using a biophysical model for robust and accelerated reconstruction of quantitative, artifact‐free and denoised images

Torop

Kothapalli

Sun

et al. 2020

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

Purpose To introduce a novel deep learning method for Robust and Accelerated Reconstruction (RoAR) of quantitative and B0‐inhomogeneity‐corrected R2* maps from multi‐gradient recalled echo (mGRE) MRI data. Methods RoAR trains a convolutional neural network (CNN) to generate quantitative R2∗ maps free from field inhomogeneity artifacts by adopting a self‐supervised learning strategy given (a) mGRE magnitude images, (b) the biophysical model describing mGRE signal decay, and (c) preliminary‐evaluated F‐function accounting for contribution of macroscopic B0 field inhomogeneities. Importantly, no ground‐truth R2* images are required and F‐function is only needed during RoAR training but not application. Results We show that RoAR preserves all features of R2* maps while offering significant improvements over existing methods in computation speed (seconds vs. hours) and reduced sensitivity to noise. Even for data with SNR = 5 RoAR produced R2* maps with accuracy of 22% while voxel‐wise analysis accuracy was 47%. For SNR = 10 the RoAR accuracy increased to 17% vs. 24% for direct voxel‐wise analysis. Conclusions RoAR is trained to recognize the macroscopic magnetic field inhomogeneities directly from the input magnitude‐only mGRE data and eliminate their effect on R2∗ measurements. RoAR training is based on the biophysical model and does not require ground‐truth R2* maps. Since RoAR utilizes signal information not just from individual voxels but also accounts for spatial patterns of the signals in the images, it reduces the sensitivity of R2* maps to the noise in the data. These features plus high computational speed provide significant benefits for the potential usage of RoAR in clinical settings.

show abstract

Image Restoration Using Total Variation Regularized Deep Image Prior

Cited by 136 publications

References 31 publications

Diffraction tomography with a deep image prior

Diffraction tomography with a deep image prior

Sub-Dip: Optimization On A Subspace With Deep Image Prior Regularization And Application To Superresolution

Deep learning using a biophysical model for robust and accelerated reconstruction of quantitative, artifact‐free and denoised images

Contact Info

Product

Resources

About