LDRD: Smart Acquisition and Sampling for Simulation and Imaging

Event Sponsor: 
Argonne Leadership Computing Facility Seminar
Start Date: 
Aug 15 2018 - 1:00pm
Building/Room: 
Building 240/Room 4301
Location: 
Argonne National Laboratory
Speaker(s): 
Yan Zhang
Speaker(s) Title: 
Postdoctoral Appointee, ANL
Host: 
Venkat Vishwanath & Elise Jennings

Novel data acquisition schemes have been an emerging need for scanning microscopy-based imaging techniques to reduce the time in data acquisition and to minimize probing radiation in sample exposure. Varies sparse sampling schemes have been studied and are ideally suited for such applications where the images can be reconstructed from a sparse set of measurements. Dynamic sparse sampling methods, particularly supervised learning based iterative sampling algorithms, have shown promising results for sampling pixel locations on the edges or boundaries during imaging. Here, w e address this issue in three different aspects which leads to three applications. First, we present a novel machine learning based method for dynamic sparse sampling of EDS data using a scanning electron microscope. Our method, based on the supervised learning approach for dynamic sampling algorithm and neural networks-based classification of EDS data, allows a dramatic reduction in the total sampling of up to 90%, while maintaining the fidelity of the reconstructed elemental maps and spectroscopic data. We believe this approach will enable imaging and elemental mapping of materials that would otherwise be inaccessible to these analysis techniques. Second, we improve out method and develop SLADS-Net in which the training images either have similar information content or completely different information content, when compare to the testing images. We observe that deep neural network-based training results in superior performance when the training and testing images are not s imilar. We also discuss the development of a pre-trained SLADS-Net that uses generic images for training. Here, the neural network parameters are pre-trained so that users can directly apply SLADS-Net for imaging experiments. Last, we extend dynamic sampling method for imaging skeleton-like objects such as metal dendrites. We address a new unsupervised learning approach using Hierarchical Gaussian Mixture Models (HGMM) to dynamically sample metal dendrites. This technique is very useful if the users are interested in fast imaging the primary and secondary arms of metal dendrites and provides an alternate when boundary-focused dynamic sampling is not applicable in materials imaging.
 

Miscellaneous Information: 

This seminar will be streamed, see details at https://anlpress.cels.anl.gov/mcs-streaming-seminars.

Please click here [schedule.ics] to add this event to your calendar.