Low-complexity convolutional neural networks for automatic target recognition

Over the decades, several algorithms have been proposed for designing automatic target recognition systems based on synthetic aperture radar imagery. Recently, with the rise of Deep Learning, there has been growing interest in developing neural network based automatic target recognition systems for synthetic aperture radar applications. However, these networks are typically complex in terms of storage and computation which inhibits their deployment in the field, where such resources are heavily constrained. In order to reduce the cost of implementing these networks, in this thesis we develop a set of compact network architectures and train them in fixed-point. Our proposed method achieves an overall 984× reduction in terms of storage requirements and 71× reduction in terms of computational complexity compared to state-of-the-art convolutional neural networks for automatic target recognition, while maintaining a classification accuracy of >99% on the MSTAR dataset.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Dbouk, Hassan

Contributors dc:contributor

• Shanbhag, Naresh R

• deep learning
• neural networks
• automatic target recognition
• synthetic aperture radar
• quantization

Statement dc:rights

• Copyright 2020 Hassan Dbouk

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108091

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/108091