Texture Content-Based Retrieval Using Content Descriptions

Kuan, Joseph K. P., Joyce, Dan W. and Lewis, Paul H. (1999) Texture Content-Based Retrieval Using Content Descriptions. Storage and Retrieval for Image and Video Databases VII. 75--85 .

Abstract

We have developed a content based retrieval scheme for texture which makes use of text based descriptions of the texture as well as explicit representations of the image content. The texture representation technique is based on our previous work which uses very simple texture primitives such as edges and plain regions to generate features. Other texture representations that apply more complicated statistics can be difficult to transcribe into understandable forms for normal users. With the simplicity of our features, we can express the texture in terms of simple language. This gives the user a more flexible interface to the image texture and attempts to bridge the gap between semantics and computed features. A number of benefits can be achieved, offering new possibilities for content based retrieval with texture. For example, the user can request a texture image without necessarily knowing what types of textures are stored and can modify a pre-existing texture using high level instructions. In this paper we describe the method of translating such features and the partial weighted Euclidean distance matching which allows users to describe only the aspects of the texture in which they are interested. This allows them to gradually refine their texture descriptions.

This record has no associated files available for download.

Contributors

Download statistics