Convolutional Neural Networks for 1-D Many-Channel Data

Report No. ARL-TR-8372
Authors: John S Hyatt, Eliseo Iglesias, and Michael Lee
Date/Pages: June 2018; 44 pages
Abstract: Deep convolutional neural networks (CNNs) represent the state of the art in image recognition. The same properties that led to their success in that domain allow them to be applied to superficially very different problems with minimal modification. In this work, we have modified a simple CNN, originally written to classify digits in the MNIST database (28 x 28 pixels, 1 channel), for use on 1-D acoustic data taken from experiments focused on crack detection (8,000 data points, 72 channels). Though the model's predictive ability is limited to fitting the trend, its partial success suggests that the application of convolutional networks to novel domains deserves further attention.
Distribution: Approved for public release
  Download Report ( 1.651 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.
 

Last Update / Reviewed: June 1, 2018