Previous [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

Journal of Inforamtion Science and Engineering, Vol.17 No.5, pp.729-751 (September 2001)

Color Image Retrieval Using Geometric Properties

Ing-Sheen Hsieh and Kuo-Chin Fan*
Institute of Computer Science and Information Engineering
National Central University
Chungli, 320 Taiwan
*Department of Engineering Science
National Cheng Kung University
Chungli, 320 Taiwan
*E-mail: kcfan@csie.ncu.edu.tw

In this paper, we present a novel region-based color image retrieval system using geometric properties. First, a region-growing technique is employed to cluster the connected color pixels with the same color in an image to form color regions, which are the primitive elements in our proposed approach. In the feature extraction module, two important descriptive geometry features are extracted, the spatial relational graph (SRG) and the Fourier description coefficients (FDCs) of each color region. In the matching module, relational distance graph matching between two SRGs is performed to find the best matches with the minimum relational distance. Then, shape matching is applied to obtain the best match with the minimum geometric distance. In our work, the method proposed by Cinque [1] is modified to perform the relational distance graph matching, then the wavelet transform is applied to extract the critical points on the contour of color regions. Experimental results reveal the feasibility of our proposed approach in solving the color image retrieval problem.

Keywords: image retrieval, shape matching, maximin algorithm, wavelet transform, Hungarian method

Full Text () Retrieve PDF document (200109_02.pdf)

Received April 21, 1999; revised May 31 & August 29, 2000; accepted October 5, 2000.
Communicated by Arbee L. P. Chen.