Image retrieval using texture based on DCT
Hee-Jung Bae, Sunghwan Jung · 2002
A new content-based image retrieval method using texture information is proposed. For efficient image retrieval, it extracts low-level image texture features as a content retrieval index. The proposed feature extraction uses some discrete cosine transform (DCT) coefficients which represent some dominant directions and gray level variations of the image. Our method also uses a hierarchical similarity measure for efficient retrieval, making it possible to reduce the search space in a large image database. In an experiment using a database of 200 images, our method shows a higher performance in the retrieval rate and processing time than Tamura's (1976). Therefore, our approach for extracting features in the transform domain can provide a solution to storage space problem with its ease of computation.