scholarly journals Compressive Sensing Spectroscopy Using a Residual Convolutional Neural Network

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 594 ◽  
Author(s):  
Cheolsun Kim ◽  
Dongju Park ◽  
Heung-No Lee
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Compressive Sensing ◽  
Prior Information ◽  
A Priori ◽  
Sparse Recovery ◽  
Mean Squared Errors ◽  
Input Spectrum ◽  
Spectral Recovery ◽  
Recovery Methods

Compressive sensing (CS) spectroscopy is well known for developing a compact spectrometer which consists of two parts: compressively measuring an input spectrum and recovering the spectrum using reconstruction techniques. Our goal here is to propose a novel residual convolutional neural network (ResCNN) for reconstructing the spectrum from the compressed measurements. The proposed ResCNN comprises learnable layers and a residual connection between the input and the output of these learnable layers. The ResCNN is trained using both synthetic and measured spectral datasets. The results demonstrate that ResCNN shows better spectral recovery performance in terms of average root mean squared errors (RMSEs) and peak signal to noise ratios (PSNRs) than existing approaches such as the sparse recovery methods and the spectral recovery using CNN. Unlike sparse recovery methods, ResCNN does not require a priori knowledge of a sparsifying basis nor prior information on the spectral features of the dataset. Moreover, ResCNN produces stable reconstructions under noisy conditions. Finally, ResCNN is converged faster than CNN.

White noise attenuation of seismic trace by integrating variational mode decomposition with convolutional neural network

Geophysics ◽  
2019 ◽  
Vol 84 (5) ◽  
pp. V307-V317 ◽  
Author(s):  
Hao Wu ◽  
Bo Zhang ◽  
Tengfei Lin ◽  
Fangyu Li ◽  
Naihao Liu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
White Noise ◽  
Seismic Data ◽  
Prior Information ◽  
Training Procedure ◽  
Noise Attenuation ◽  
Variational Mode Decomposition ◽  
Time Frequency ◽  
Mode Decomposition

Seismic noise attenuation is an important step in seismic data processing. Most noise attenuation algorithms are based on the analysis of time-frequency characteristics of the seismic data and noise. We have aimed to attenuate white noise of seismic data using the convolutional neural network (CNN). Traditional CNN-based noise attenuation algorithms need prior information (the “clean” seismic data or the noise contained in the seismic) in the training process. However, it is difficult to obtain such prior information in practice. We assume that the white noise contained in the seismic data can be simulated by a sufficient number of user-generated white noise realizations. We then attenuate the seismic white noise using the modified denoising CNN (MDnCNN). The MDnCNN does not need prior clean seismic data nor pure noise in the training procedure. To accurately and efficiently learn the features of seismic data and band-limited noise at different frequency bandwidths, we first decomposed the seismic data into several intrinsic mode functions (IMFs) using variational mode decomposition and then apply our denoising process to the IMFs. We use synthetic and field data examples to illustrate the robustness and superiority of our method over the traditional methods. The experiments demonstrate that our method can not only attenuate most of the white noise but it also rejects the migration artifacts.

scholarly journals A structure-based deep learning framework for protein engineering

2019 ◽  
Author(s):  
Raghav Shroff ◽  
Austin W. Cole ◽  
Barrett R. Morrow ◽  
Daniel J. Diaz ◽  
Isaac Donnell ◽  
...  
Keyword(s):  
Neural Network ◽  
Amino Acids ◽  
Deep Learning ◽  
Protein Engineering ◽  
Convolutional Neural Network ◽  
State Of The Art ◽  
A Priori ◽  
Novel Proteins ◽  
Learning Framework

AbstractWhile deep learning methods exist to guide protein optimization, examples of novel proteins generated with these techniques require a priori mutational data. Here we report a 3D convolutional neural network that associates amino acids with neighboring chemical microenvironments at state-of-the-art accuracy. This algorithm enables identification of novel gain-of-function mutations, and subsequent experiments confirm substantive phenotypic improvements in stability-associated phenotypes in vivo across three diverse proteins.

A new fault diagnosis method based on convolutional neural network and compressive sensing

2019 ◽  
Vol 33 (11) ◽  
pp. 5177-5188 ◽  
Author(s):  
Yunfei Ma ◽  
Xisheng Jia ◽  
Huajun Bai ◽  
Guozeng Liu ◽  
Guanglong Wang ◽  
...  
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Compressive Sensing ◽  
Diagnosis Method

Deep Convolutional Neural Network for Compressive Sensing of Magnetic Resonance Images

2021 ◽  
Author(s):  
Hong Lu ◽  
Xiaofei Zou ◽  
Longlong Liao ◽  
Kenli Li ◽  
Jie Liu
Keyword(s):  
Neural Network ◽  
Magnetic Resonance ◽  
Convolutional Neural Network ◽  
Compressive Sensing ◽  
Network Architecture ◽  
Structural Similarity ◽  
Magnetic Resonance Images ◽  
Deep Convolutional Neural Network ◽  
Mr Images ◽  
Under Sampling

Compressive Sensing for Magnetic Resonance Imaging (CS-MRI) aims to reconstruct Magnetic Resonance (MR) images from under-sampled raw data. There are two challenges to improve CS-MRI methods, i.e. designing an under-sampling algorithm to achieve optimal sampling, as well as designing fast and small deep neural networks to obtain reconstructed MR images with superior quality. To improve the reconstruction quality of MR images, we propose a novel deep convolutional neural network architecture for CS-MRI named MRCSNet. The MRCSNet consists of three sub-networks, a compressive sensing sampling sub-network, an initial reconstruction sub-network, and a refined reconstruction sub-network. Experimental results demonstrate that MRCSNet generates high-quality reconstructed MR images at various under-sampling ratios, and also meets the requirements of real-time CS-MRI applications. Compared to state-of-the-art CS-MRI approaches, MRCSNet offers a significant improvement in reconstruction accuracies, such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM). Besides, it reduces the reconstruction error evaluated by the Normalized Root-Mean-Square Error (NRMSE). The source codes are available at https://github.com/TaihuLight/MRCSNet .

Sign in / Sign up

Export Citation Format

Share Document