Please use this identifier to cite or link to this item: https://hdl.handle.net/11681/37618
Title: Understanding state-of-the-art material classification through deep visualization
Authors: Donovan, Jordan T.
Keywords: Neural networks (Computer science)
Machine learning
Computer vision
Image processing
Digital images--Deconvolution
Deep visualization
Image segmentation
Object detection
Material classification
Publisher: Engineer Research and Development Center (U.S.)
Series/Report no.: Miscellaneous Paper (Information Technology Laboratory (U.S.)) ; no. ERDC/ITL MP-20-1
Abstract: Neural networks (NNs) excel at solving several complex, non-linear problems in the area of supervised learning. A prominent application of these networks is image classification. Numerous improvements over the last few decades have improved the capability of these image classifiers. However, neural networks are still a black-box for solving image classification and other sophisticated tasks. A number of experiments conducted look into exactly how neural networks solve these complex problems. This paper dismantles the neural network solution, incorporating convolution layers, of a specific material classifier. Several techniques are utilized to investigate the solution to this problem. These techniques look at specifically which pixels contribute to the decision made by the NN as well as a look at each neuron’s contribution to the decision. The purpose of this investigation is to understand the decision-making process of the NN and to use this knowledge to suggest improvements to the material classification algorithm.
Description: Miscellaneous Paper
Gov't Doc #: ERDC/ITL MP-20-1
Rights: Approved for Public Release; Distribution is Unlimited
URI: https://hdl.handle.net/11681/37618
http://dx.doi.org/10.21079/11681/37618
Appears in Collections:Miscellaneous Paper

Files in This Item:
File Description SizeFormat 
ERDC-ITL MP-20-1.pdfERDC/ITL MP-20-127.53 MBAdobe PDFThumbnail
View/Open