A mixed-scale dense convolutional neural network for image analysis

Pelt, Daniël M.; Sethian, James A.

doi:10.1073/pnas.1715832114

Title: A mixed-scale dense convolutional neural network for image analysis

Abstract

We report that deep convolutional neural networks have been successfully applied to many image-processing problems in recent works. Popular network architectures often add additional operations and connections to the standard architecture to enable training deeper networks. To achieve accurate results in practice, a large number of trainable parameters are often required. Here, we introduce a network architecture based on using dilated convolutions to capture features at different image scales and densely connecting all feature maps with each other. The resulting architecture is able to achieve accurate results with relatively few parameters and consists of a single set of operations, making it easier to implement, train, and apply in practice, and automatically adapts to different problems. Lastly, we compare results of the proposed network architecture with popular existing architectures for several segmentation problems, showing that the proposed architecture is able to achieve accurate results with fewer parameters, with a reduced risk of overfitting the training data.

Authors:

Pelt, Daniël M. ^[1]; Sethian, James A. ^[2]

Center for Applied Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720,
Center for Applied Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory, Berkeley, CA 94720,, Department of Mathematics, University of California, Berkeley, CA 94720

Publication Date:: 2017-12-26

Research Org.:: Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)

Sponsoring Org.:: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE Office of Science (SC), Basic Energy Sciences (BES); USDOE Office of Science (SC), Biological and Environmental Research (BER)

OSTI Identifier:: 1414877

Alternate Identifier(s):: OSTI ID: 1485062

Grant/Contract Number:: AC03-76SF00098; AC02-05CH11231

Resource Type:: Published Article

Journal Name:: Proceedings of the National Academy of Sciences of the United States of America

Additional Journal Information:: Journal Name: Proceedings of the National Academy of Sciences of the United States of America Journal Volume: 115 Journal Issue: 2; Journal ID: ISSN 0027-8424

Publisher:: Proceedings of the National Academy of Sciences

Country of Publication:: United States

Language:: English

Subject:: 97 MATHEMATICS AND COMPUTING; image segmentation; machine learning; convolution neural networks

Citation Formats


                    Pelt, Daniël M., and Sethian, James A.. A mixed-scale dense convolutional neural network for image analysis.  United States: N. p., 2017. 
Web.  doi:10.1073/pnas.1715832114.

Copy to clipboard


                    Pelt, Daniël M., & Sethian, James A.. A mixed-scale dense convolutional neural network for image analysis.  United States.  https://doi.org/10.1073/pnas.1715832114

Copy to clipboard


                    Pelt, Daniël M., and Sethian, James A.. Tue .  
"A mixed-scale dense convolutional neural network for image analysis".  United States.  https://doi.org/10.1073/pnas.1715832114.

Copy to clipboard


                    
@article{osti_1414877,

  title        = {A mixed-scale dense convolutional neural network for image analysis},

  author       = {Pelt, Daniël M. and Sethian, James A.},

  abstractNote = {We report that deep convolutional neural networks have been successfully applied to many image-processing problems in recent works. Popular network architectures often add additional operations and connections to the standard architecture to enable training deeper networks. To achieve accurate results in practice, a large number of trainable parameters are often required. Here, we introduce a network architecture based on using dilated convolutions to capture features at different image scales and densely connecting all feature maps with each other. The resulting architecture is able to achieve accurate results with relatively few parameters and consists of a single set of operations, making it easier to implement, train, and apply in practice, and automatically adapts to different problems. Lastly, we compare results of the proposed network architecture with popular existing architectures for several segmentation problems, showing that the proposed architecture is able to achieve accurate results with fewer parameters, with a reduced risk of overfitting the training data.},

  doi          = {10.1073/pnas.1715832114},

  journal      = {Proceedings of the National Academy of Sciences of the United States of America},

  number       = 2,

  volume       = 115,

  place        = {United States},

  year         = {2017},

  month        = {12}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Publisher's Version of Record
https://doi.org/10.1073/pnas.1715832114

Other availability

Search WorldCat to find libraries that may hold this journal

Citation Metrics:

Cited by: 125 works

Citation information provided by
Web of Science

Figures / Tables:

Fig. 1: A schematic representation of a two-layer CNN with input x, output y, and feature maps z₁ and z₂. Arrows represent convolutions with nonlinear activation.

All figures and tables (9 total)

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

Superparsing: Scalable Nonparametric Image Parsing with Superpixels
journal, October 2012

Tighe, Joseph; Lazebnik, Svetlana
International Journal of Computer Vision, Vol. 101, Issue 2
DOI: 10.1007/s11263-012-0574-z

PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation
journal, March 2012

Klöckner, Andreas; Pinto, Nicolas; Lee, Yunsup
Parallel Computing, Vol. 38, Issue 3
DOI: 10.1016/j.parco.2011.09.001

ImageNet: A large-scale hierarchical image database
conference, June 2009

Deng, Jia; Dong, Wei; Socher, Richard
2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition
DOI: 10.1109/CVPR.2009.5206848

Deep learning
journal, May 2015

LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
Nature, Vol. 521, Issue 7553
DOI: 10.1038/nature14539

Fully Convolutional Networks for Semantic Segmentation
journal, April 2017

Shelhamer, Evan; Long, Jonathan; Darrell, Trevor
IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, Issue 4
DOI: 10.1109/TPAMI.2016.2572683

Semantic object classes in video: A high-definition ground truth database
journal, January 2009

Brostow, Gabriel J.; Fauqueur, Julien; Cipolla, Roberto
Pattern Recognition Letters, Vol. 30, Issue 2
DOI: 10.1016/j.patrec.2008.04.005

Generalizing the Hough transform to detect arbitrary shapes
journal, January 1981

Ballard, D. H.
Pattern Recognition, Vol. 13, Issue 2
DOI: 10.1016/0031-3203(81)90009-1

Radio frequency interference mitigation using deep convolutional neural networks
journal, January 2017

Akeret, J.; Chang, C.; Lucchi, A.
Astronomy and Computing, Vol. 18
DOI: 10.1016/j.ascom.2017.01.002

Image-to-Image Translation with Conditional Adversarial Networks
conference, July 2017

Isola, Phillip; Zhu, Jun-Yan; Zhou, Tinghui
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
DOI: 10.1109/CVPR.2017.632

Caffe: Convolutional Architecture for Fast Feature Embedding
conference, January 2014

Jia, Yangqing; Shelhamer, Evan; Donahue, Jeff
Proceedings of the ACM International Conference on Multimedia - MM '14
DOI: 10.1145/2647868.2654889

Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique
journal, July 2001

Yongbum Lee, ; Hara, T.; Fujita, H.
IEEE Transactions on Medical Imaging, Vol. 20, Issue 7
DOI: 10.1109/42.932744

Figures / Tables found in this record:

Figures/Tables have been extracted from DOE-funded journal article accepted manuscripts.

Similar Records in DOE PAGES and OSTI.GOV collections:

Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices

Dataset Churchill, R M ; the DIII-D team

The multi-scale, mutli-physics nature of fusion plasmas makes predicting plasma events challenging. Recent advances in deep convolutional neural network architectures (CNN) utilizing dilated convolutions enable accurate predictions on sequences which have long-range, multi-scale characteristics, such as the time-series generated by diagnostic instruments observing fusion plasmas. Here we apply this neural network architecture to the popular problem of disruption prediction in fusion tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. ECEi measures a fundamental plasma quantity (electron temperature) with high temporal resolution over the entire plasma discharge, making it sensitive to a number of potential pre-disruptions markers with different temporal and spatial scales. Promising, initial disruption prediction results are obtained training a deep CNN with large receptive field ({more »« less
https://doi.org/10.11578/1661171

View Dataset
Application of convolutional neural networks for stellar spectral classification

Journal Article Sharma, Kaushal ; Kembhavi, Ajit ; Kembhavi, Aniruddha ; ... - Monthly Notices of the Royal Astronomical Society

ABSTRACT Due to the ever-expanding volume of observed spectroscopic data from surveys such as SDSS and LAMOST, it has become important to apply artificial intelligence (AI) techniques for analysing stellar spectra to solve spectral classification and regression problems like the determination of stellar atmospheric parameters Teff,more »« less
https://doi.org/10.1093/mnras/stz3100
A data-driven CO₂ leakage detection using seismic data and spatial–temporal densely connected convolutional neural networks

Journal Article Zhou, Zheng ; Lin, Youzuo ; Zhang, Zhongping ; ... - International Journal of Greenhouse Gas Control

In carbon capture and sequestration (also known as carbon capture and storage, or CCS), developing effective monitoring methods is needed to detect and respond to CO₂ leakage. CO₂ leakage detection methods rely on geophysical observations and monitoring sensor network. However, traditional methods usually require the development of site-specific physical models and expert interpretation, and the effectiveness of these methods can be limited to different application locations, operational scenarios, and conditions. Here, we developed a novel data-driven leakage detection method based on densely connected convolutional neural networks. Our method is an end-to-end detection approach, that differs from conventional leakage monitoring methodsmore »« less
Cited by 11
https://doi.org/10.1016/j.ijggc.2019.102790

Full Text Available
Deep convolutional neural networks for multi-scale time-series classification and application to tokamak disruption prediction using raw, high temporal resolution diagnostic data

Journal Article Churchill, R. M. ; Tobias, B. ; Zhu, Y. - Physics of Plasmas

In this paper, we discuss recent advances in deep convolutional neural networks (CNNs) for sequence learning, which allow identifying longrange, multi-scale phenomena in long sequences, such as those found in fusion plasmas. We point out several benefits of these deep CNN architectures, such as not requiring experts such as physicists to hand-craft input data features, the ability to capture longer range dependencies compared to the more common sequence neural networks (recurrent neural networks like long short-term memory networks), and the comparative computational efficiency. We apply this neural network architecture to the popular problem of disruption prediction in fusion energy tokamaks,more »« less
Cited by 28
https://doi.org/10.1063/1.5144458

Full Text Available
Data-driven modeling of coarse mesh turbulence for reactor transient analysis using convolutional recurrent neural networks

Journal Article Liu, Yang ; Hu, Rui ; Kraus, Adam R. ; ... - Nuclear Engineering and Design

Advanced nuclear reactors often exhibit complex thermal-fluid phenomena during transients. To accurately capture such phenomena, a coarse-mesh three-dimensional (3-D) modeling capability is desired for modern nuclear-system code. In the coarse-mesh 3-D modeling of advanced-reactor transients that involve flow and heat transfer, accurately predicting the turbulent viscosity is a challenging task that requires an accurate and computationally efficient model to capture the unresolved fine-scale turbulence. In this work, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor. The model has a coarse mesh setupmore »« less
https://doi.org/10.1016/j.nucengdes.2022.111716

Full Text Available

Similar Records

Title: A mixed-scale dense convolutional neural network for image analysis

Abstract

Citation Formats

Figures / Tables:

Superparsing: Scalable Nonparametric Image Parsing with Superpixels journal, October 2012

PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation journal, March 2012

ImageNet: A large-scale hierarchical image database conference, June 2009

Deep learning journal, May 2015

Fully Convolutional Networks for Semantic Segmentation journal, April 2017

Semantic object classes in video: A high-definition ground truth database journal, January 2009

Generalizing the Hough transform to detect arbitrary shapes journal, January 1981

Radio frequency interference mitigation using deep convolutional neural networks journal, January 2017

Image-to-Image Translation with Conditional Adversarial Networks conference, July 2017

Caffe: Convolutional Architecture for Fast Feature Embedding conference, January 2014

Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique journal, July 2001

Superparsing: Scalable Nonparametric Image Parsing with Superpixels
journal, October 2012

PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation
journal, March 2012

ImageNet: A large-scale hierarchical image database
conference, June 2009

Deep learning
journal, May 2015

Fully Convolutional Networks for Semantic Segmentation
journal, April 2017

Semantic object classes in video: A high-definition ground truth database
journal, January 2009

Generalizing the Hough transform to detect arbitrary shapes
journal, January 1981

Radio frequency interference mitigation using deep convolutional neural networks
journal, January 2017

Image-to-Image Translation with Conditional Adversarial Networks
conference, July 2017

Caffe: Convolutional Architecture for Fast Feature Embedding
conference, January 2014

Automated detection of pulmonary nodules in helical CT images based on an improved template-matching technique
journal, July 2001