An Improved Approach Based on FCM Using Feature Fusion for Medical Image Retrieval

Herein we describe a system for Gastroscopic image which supports image retrieval by content. Attention is focused on content-based approaches for the efficient representation and retrieval of medical images. A new improved image retrieval method with more robustness based on image entropy is proposed, which can be defined as spatial weighted entropy. Then a feature fusion algorithm of medical image retrieval based on FCM using spatial weighted entropy combining with color correlogram is presented. A prototype system which supports query by example is designed and implemented on windows 2003 operating system, using VC# and SQL server 2000.The performance of the method is illustrated using examples from an image database composed of 1361 Gastroscopic images, and the comparison of the retrieval results shows that the approach proposed in this paper is effective.