Random walks based segmentation approach for image retrieval

In this paper a region-based image indexing and retrieval (RBIR) algorithm is presented. As a basis for the indexing, a novel spectral segmentation approach using random walks on graphs is introduced. Based on the extracted regions, characteristic features are estimated using color and texture information. The focus of this study is to improve the capture of regions so as to enhance indexing and retrieval performance. An implemented prototype system has demonstrated a promising retrieval performance for a test database containing more than 1000 color algae images.