Channel Color Energy Feature Representing Color and Texture in Content-Based Image Retrieval
Jung Jae Woong, Kwon Tae Wan, Park Seop Hyeong · Journal of the Institute of Electronics Engineers of Korea · 2004
In the field of content-based image retrieval, many numerical features have been proposed for representing visual image content such as color, torture, and shape. Because the features are assumed to be independent, each of them is extracted without ny consideration of the others. In this paper, we consider the relationship between color and texture and propose a new feature called CCE(channel color energy). Simulation results with natural images show that the proposed method outperforms the conventional regular weighted comparison method and SCFT(sequential chromatic Fourier transform)-based color torture method.