Abstract

In this paper, we propose a novel one- and multi-dimensional signal classification neural network system that employs a set of criteria extracted from the signal representation in different transform domains, denoted the multicriteria multitransform neural network classifier. The signal projection, in each appropriately selected transform domain, reveals unique signal characteristics. The criteria in the different domains are properly formulated and their parameters adapted to obtain classification with desirable implementation properties such as speed and accuracy. Results for image classification confirm the improved classification performance relative to existing techniques. In addition to the improved computational efficiency and accuracy, preliminary results indicate that the proposed technique lends itself to higher classification accuracy in the presence of additive noise.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call