Asia-Pacific Journal of Neural Networks and Its Applications
| Volume 2, No. 2, 2018, pp 1-6 | ||
| Abstract |
Active Sonar Target Classification Using Multi-aspect based Sensing and Deep Belief Network
|
|
Abstract
This paper deals with multi-aspect based sensing scheme for active sonar target classification to improve the classification performance using the synthesized active sonar returns and deep belief network. Classification experiment was performed based on single and multi-aspect schemes. As a result, the proposed multi-aspect based scheme showed better performances than single aspect based scheme and conventional backpropagation neural network classifier.