An Intelligent On-line System for Content Based Image Retrieval

Abstract

This study builds on our previous work and proposes an intelligent content-based picture retrieval system [1]. The foundation
of the system is our neuro-fuzzy method, which is detailed in [1]. For example, you may enter “mostly red & many blue &
few green” into the CBIR system online, and it will return photos with their secret values as an output. The system learns the
meaning of mainly red, many red, and few red using a neural network, and it employs fuzzy logic to understand many natural
phrases like mostly, many, and few. The C programming language and CGI scripts were used to implement the system on the
World Wide Web. We have shown some first findings with a single query in [1]. Expanding our online system to handle numerous
requests, such “many blue and few green,” is the focus of this research. Our online system was used to perform many
experiments, the results of which are presented and discussed in this study.