Deep BCD-Net Using Identical Encoding-Decoding CNN Structures for Iterative Image Recovery

02/20/2018
by   Il Yong Chun, et al.
0

In "extreme" computational imaging that collects extremely undersampled or noisy measurements, obtaining an accurate image within a reasonable computing time is challenging. Incorporating image mapping convolutional neural networks (CNN) to iterative image recovery has great potential to resolve this issue. This paper 1) incorporates image mapping CNN using identical convolutional kernels in both encoders and decoders into block coordinate descent (BCD) optimization method -- referred to BCD-Net using identical encoding-decoding CNN structures -- and 2) applies alternating direction method of multipliers to train the proposed BCD-Net. Numerical experiments show that, for a) denoising moderately low signal-to-noise-ratio images and b) extremely undersampled magnetic resonance imaging, the proposed BCD-Net achieves (significantly) more accurate image recovery, compared to BCD-Net using distinct encoding-decoding structures and/or the conventional image recovery model using both wavelets and total variation.

READ FULL TEXT

page 3

page 4

research
12/02/2020

An Improved Iterative Neural Network for High-Quality Image-Domain Material Decomposition in Dual-Energy CT

Dual-energy computed tomography (DECT) has been widely used in many appl...
research
07/30/2018

Efficient feature learning and multi-size image steganalysis based on CNN

For steganalysis, many studies showed that convolutional neural network ...
research
06/05/2019

Improved low-count quantitative PET reconstruction with a variational neural network

Image reconstruction in low-count PET is particularly challenging becaus...
research
01/16/2018

An Accurate and Real-time Self-blast Glass Insulator Location Method Based On Faster R-CNN and U-net with Aerial Images

The location of broken insulators in aerial images is a challenging task...
research
10/25/2020

Applying convolutional neural networks to extremely sparse image datasets using an image subdivision approach

Purpose: The aim of this work is to demonstrate that convolutional neura...
research
06/25/2022

Sequential image recovery using joint hierarchical Bayesian learning

Recovering temporal image sequences (videos) based on indirect, noisy, o...

Please sign up or login with your details

Forgot password? Click here to reset