Visual data indexing and retrieval using color feature

Recent advances in visual data storage and retrieval technologies have made the creation of very large databases feasible. Color indexing is one of the crucial issues in the management of color image and video databases. In this dissertation we investigate an approach based on the color feature extraction and indexing of images for the purpose of content-based color image and video database retrieval. Since the high computational complexity has been one of the main barriers towards the use of similarity measures, such as histogram intersection distance, in very large database, we present a hierarchical automatic color indexing scheme and used to subset images before more sophisticated techniques are applied for precise retrieval. The use of automatically indexed color contents of images as filtering and matching features in a hierarchical scheme is studied with full algorithm implementation procedure described in detail. In the meantime, we developed a set of programs to extract the color features from which the color vector complex of images is derived for accomplishing both color image database indexing and video sequence parsing. The discrimination between different image scenes based on color vector difference measurement is also studied. The experimental results demonstrate that our approach is the state-of-art of work with high efficiency and low computation complexity. Through applying it to visual data we can expect getting not only a way of color image discrimination but also a way of color video sequence segmentation or shot boundary detection based on color features.