A new method of image data fusion based on FNN

The image data are obtained by a variety of multimedia, information equipment, which include amount of information, extensive coverage, and redundancy in ubiquitous computing paradigm. In order to make use of these information reasonably and efficiently, it is necessary to fuse such massive data, therefore the multi-sensor image information fusion become a key technology to ubiquitous computing. A new approach of image data fusion based on fuzzy neural network (FNN) is presented in this paper. Here, neural network is a parallel information processing model, there are more adaptive and self-organization, and can accomplish the complexity of real-time computing and mass data retrieval, it demonstrate its unique superiority of image understanding, pattern recognition and the handling of incomplete information. In the new approach, fuzzy theory has the advantages such as: easy-to-understand, flexible and inclusive of non-precise data, as well as using expert knowledge and based on the natural language. The fuzzy neural network system developed by us has used in multi-sensor image fusion, proved that the fusion is fast, effective, good, and can meet the real-time requirements of ubiquitous computing.

1. Introduction

The rapid development of Computers and Internet technology has given birth to a new computing model-Pervasive Computing. Ubiquitous computing is a way that can calculate on when and where, as long as necessary, can access the required information on some devices. Ubiquitous computing brings convenience to the people, at the same time it also brings new challenges to the data-processing technology^[1][2]. As the image information get in the environment of ubiquitous computing is very complex, including multi-platform, multi-temporal, multi-sensor, multi-resolution images, and large-scale, high-precision, real-time multi-source remote sensing data obtained by using optical, infrared, radar remote sensing and other optical measurement system, in order to meet real-time requirements, the problem how to fusion the image data from multi-sensor rapidly and effectively has become urgent^[3].