Effective invariant features for shape-based image retrieval: Research Articles
Journal of the American Society for Information Science and Technology
| Hi-index | 0.00 |
Content-based retrievals in image databases require appropriatefeatures of images to be derived and used in theindexing and searching process. Fast and accurate retrievalsare crucial from a user point of view. The kindsof features used, their organization in suitable data structures,and the similarity search scheme, directly affect thespeed and quality of content-based retrievals. In this paper,we evaluate the relative significance of three differentimage features - geometry, moments, and Fourier descriptorsin the context of content-based retrievals, and presentexperimental results. The evaluation metrics are retrievalquality and search time. These could be used to tune thesearch process which speeds up the retrieval without significantlyaffecting the retrieval quality.