Image Retrieval Using Local Color Features

• Main Authors: Tzu-Chien Wu, 吳子謙
• Other Authors: Jin-Jang Leou
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/60862434592065053841

碩士 === 國立中正大學 === 資訊工程研究所 === 91 === Multimedia database systems are becoming increasingly important with the advent of broadband networks, high-powered PC’s, audio/visual compression standards, and many applications, such as digital libraries and trademark and copyright databases. In this study, an image indexing and retrieval system using local color features and the weighted color distortion measure is proposed. In the proposed system, based on illumination variation, each image is segmented into several regions by the watershed segmentation algorithm first, and the mutual relationships between connected color regions are extracted as local color features. That is, an image can be represented as a set of adjacent color regions and the mutual relationships between connected color regions. In the image retrieval stage, the similarity between a query image and a target image will contain not only direct region correspondence but also the mutual relationships between connected color regions, extracted as local color features. To obtain the better retrieval precision, a modified weighted color distortion measure is applied. Different importances of various color elements in the YUV or HSI color space will receive different color weights so that the illumination variation effect will be greatly reduced. Based on the experimental results obtained in this study, the performance of the proposed system is better than that of three existing systems for comparison. This shows the feasibitity of the proposed system.