Research on remote sensing image classification using neural network based on rough sets
Zhaocong Wu · 2002
This paper presents a new approach of remote sensing image classification based on rough BP neural networks (RBPNN), which promises to overcome some problems encountered in a conventional BP neural network (BPNN). The novelty of this network lies in applying rough sets for extracting classification rules directly from the training dataset, then there is no extra parameters had to be set for the network. While the architecture and training method of this network are presented in this article, a survey and analysis of the RBPNN for the classification of remote sensing multi-spectral images is also discussed. The successful application of this network in land cover classification illustrates the simple computation and exact accuracy of the new neural network and the flexibility and practicality of this new approach.