Comparative analysis of image retrieval approaches

Abstract

The rapid development of technologies and steadily growing amounts of digital information highlight the need of developing an accessing system. Content based image indexing and retrieval has been one of the most important research areas in computer science for the last decades. This paper investigates the approaches of content based image retrieval (CBIR) based on the low level features such as color, shape, spatial and texture analysis. It examines the shortcomings of conventional image retrieval approaches. Although CBIR offer some solutions for image retrieval in a large image database, there is still room for improvement. Due to the lack of human perception ability in image retrieval methods, their usually use low level features for image retrieval. This shortcoming of CBIR system will be analyzed by comparing it to semantic based image retrieval where low level features will be integrated with high level semantic features. © 2008 Springer-Verlag.