New Advancements in Image Segmentation for CBIR

Abstract

The process of segmenting images is one of the most critical ones in automatic image analysis whose goal can be regarded as to find what objects are presented in images (Pavlidis, 1988). Image segmentation consists of subdividing an image into its constituent parts and extracting these parts of interest (objects). A large number of segmentation algorithms have been developed since the middle of 1960’s (see survey papers and books, for example, Bovik, 2000; Fu & Mui, 1981; Lucchese & Mitra, 2001; Medioni, Lee, & Tang, 2000; Pal & Pal, 1993; Zhang, 2001), and this number continually increases from year to year in a fast rate. This number had attended, 10 years ago, the order of thousands (Zhang & Gerbrands, 1994). However, none of the proposed segmentation algorithms is generally applicable to all images, and different algorithms are not equally suitable for a particular application. Though several thousands of algorithms have been proposed, improvements for existing algorithms and developments for treating new applications are still going on.