This paper presents an effective texture image retrieval system for content-based texture image database query and retrieval. We focus on the analysis of texture and color features in the Lab color space to achieve effective texture image retrieval. Five properties of texture are computed in the “L” component in our system, i.e. coarseness, contrast, busyness, complexity and strength. These properties of texture are given conceptual definitions in terms of spatial changes in intensity. Moreover, two components of the color dimensions are also applied in our system, i.e. the “a” and “b” components. Experiment results show that our system based on texture and color features is capable of achieving more effective.
