SAIDE: Scaling Analytics for Image-based Data from Experiments

Research across science domains is increasingly reliant on image-based data from experiments. The challenge is to analyze the data torrent generated by advanced instruments in a timely manner and provide insights such as measurements for decision-making. Software tools are in high demand for scientists to uncover relevant, but hidden, information in digital images, such as those coming from new materials. A group of computational and material scientists have embraced the multi-disciplinary work of designing software applications, coordinating research efforts connecting (1) emerging algorithms for dealing with complex and large datasets; (2) data analysis methods with basis on pattern recognition and machine learning; and (3) advances in evolving computer architectures. These new trends will accelerate the analyses of image-based recordings, scaling scientific procedures by reducing time between experiments, increasing efficiency, and opening more opportunities for more users of the imaging facilities. This paper aims to provide an overview of our algorithms and deployed software tools, showing results across image scales and how each tool component plays a role in improving image understanding within DOE national laboratories.

[1]  Kunal Singha,et al.  Computer Simulations of Textile Non-Woven Structures , 2012 .

[2]  Luciano da Fontoura Costa,et al.  A texture approach to leukocyte recognition , 2004, Real Time Imaging.

[3]  Gunther H. Weber,et al.  Augmented Topological Descriptors of Pore Networks for Material Science , 2012, IEEE Transactions on Visualization and Computer Graphics.

[4]  S. Michael Spottswood,et al.  Reengineering Aircraft Structural Life Prediction Using a Digital Twin , 2011 .

[5]  Wilhelm Burger,et al.  Digital Image Processing - An Algorithmic Introduction using Java , 2016, Texts in Computer Science.

[6]  Francisco Nivando Bezerra,et al.  Multiscale Corner Detection in Planar Shapes , 2012, Journal of Mathematical Imaging and Vision.

[7]  Misha Denil,et al.  Linear and Parallel Learning of Markov Random Fields , 2013, ICML.

[8]  Arie Shoshani,et al.  Data Crosscutting Requirements Review , 2013 .

[9]  J. Sethian,et al.  Multiscale Modeling of Membrane Rearrangement, Drainage, and Rupture in Evolving Foams , 2013, Science.

[10]  Talita Perciano,et al.  Structure recognition from high resolution images of ceramic composites , 2014, 2014 IEEE International Conference on Big Data (Big Data).

[11]  Talita Perciano,et al.  Reduced-complexity image segmentation under parallel Markov Random Field formulation using graph partitioning , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[12]  Fátima N. S. de Medeiros,et al.  Wavelet Analysis for Wind Fields Estimation , 2010, Sensors.

[13]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[14]  D. Ushizima,et al.  Block Copolymer Packing Limits and Interfacial Reconfigurability in the Assembly of Periodic Mesoporous Organosilicas , 2015 .

[15]  E. Wes Bethel,et al.  Material Science Image Analysis using Quant-CT in ImageJ , 2012 .

[16]  W. Clem Karl,et al.  Learning-Based Object Identification and Segmentation Using Dual-Energy CT Images for Security , 2015, IEEE Transactions on Image Processing.

[17]  Alok Choudhary,et al.  Synergistic Challenges in Data-Intensive Science and Exascale Computing: DOE ASCAC Data Subcommittee Report , 2013 .

[18]  Drew Conway,et al.  Machine Learning for Hackers , 2012 .

[19]  Daniela Ushizima,et al.  Segmentation of subcellular compartments combining superpixel representation with Voronoi diagrams , 2015 .

[20]  R. Ritchie,et al.  Real-Time Quantitative Imaging of Failure Events in Materials under Load at Temperatures above 1700°C , 2012 .