A Theory of Multiscale, Curvature-Based Shape Representation for Planar Curves
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Comparison of Shape Retrieval Using Fourier Descriptors and Short-Time Fourier Descriptors
PCM '01 Proceedings of the Second IEEE Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
A flexible content-based image retrieval model and a customizable system for the retrieval of shapes
Journal of the American Society for Information Science and Technology
Hi-index | 0.00 |
This poster overviews the main characteristics of a flexible retrieval systems of shapes present in binary images and discusses some evaluation results. The system applies multiple indexing criteria of the shapes synthesizing distinct characteristics such as global features of the objects contour (Fourier Coefficients), boundary irregularities (Multifractal Spectrum), presence of concavities and convexities on the boundary (Contour Scale Space distribution). The system is flexible since it allows customizing the retrieval function to fit an application need. The query is a binary image containing the desired shape and a set of parameters specifying the distinct importance of the shape characteristics that must be taken into account to evaluate the relevance of the retrieved shapes. The retrieval function is then defined as a Flexible Multicriteria fusion Function producing ranked results. The evaluation experiments showed that this system can be suited to different retrieval purposes, and that generally the combination of the distinct shape indexing criteria increases both Recall and Precision with respect to the application of any single indexing criterion alone.