Contour Descriptor Based on Affine Invariance Normalization
The advent of multimedia and large image collections in different domains and applications brings with it a necessity for image retrieval systems. Image retrieval consists of the techniques used for query specification and retrieval of images from a digital image collection. It is considered part of the field of Information Retrieval (IR), a large and mature research area. IR is the field that deals with “the representation, storage, organization and access to information items.” The purpose of information retrieval is to provide the user with easy access to the information items of interest. Image retrieval has become an active research and development domain since the early 1970s . During the last decade, research on image retrieval gained high importance. The most frequent and common means for image retrieval is to index images with text keywords. Although this technique seems to be simple, it rapidly becomes laborious when faced with large volumes of images. On the other hand, images are rich in content, and this can be exploited. So to overcome difficulties due to the huge data volume, content-based image retrieval (CBIR) emerged as a promising means for retrieving images and browsing large image databases.
KeywordsImage Retrieval Planar Curve Shape Descriptor Information Item Invariance Normalization
Unable to display preview. Download preview PDF.
- 3.Cyganski D, Vaz RF (1991) A linear signal decomposition approach to affine invariant contour identification. In: Proc. of SPIE-Intelligent Robots and Computer Vision X: Algorithms and Techniques 1607:98–109Google Scholar
- 8.MPEG-7 Overview (Version 10) (2004), ISO/IEC JTC1/SC29/WG11Google Scholar