A New Probability Neural Network for Image Classification Problem

T.M. Nguyen, M.G. Sarwer, and Q.M.J. Wu (Canada)

Keywords

Probability neural network, image classification, super vised learning, and SAR image.

Abstract

Based on the extension of Bayes Classifier, we propose a new adaptive probability neural network (ProbNNet) to ap ply for image classification problem. Our proposed system could not only successfully classify the image when it is significantly degraded by high noise but also make the ef fect of noise on the final image much less. This new system has been tested with SAR image and natural image with ex cellent results.

Important Links:



Go Back