While convolutional neural networks (CNNs) are powerful tools in machine learning, their construction is far from a science. In addition, instantiations of CNNs are highly memory expensive and typically require large training sets. Wavelet scattering networks (WSNs) could provide a simple means of testing quantization schemes for CNNs, without the added complexity of adjustable parameters. Using the MSTAR database, the performance of a WSN in combination with several quantization schemes is examined.
Implementation of convolutional neural networks (CNNs) as classifiers has only recently found application in SAR multi-target classification. Despite the creation of several successful architectures, a general approach to CNN design and training has not been determined. In this paper, the basics of CNN architecture and learning algorithms are discussed. The MSTAR data set is used to demonstrate the effect of individual parameter changes to overall network performance.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.