| With the advent of digital photography and advancement in digitization process, everyday a great number of digital images are produced, resulting in a rapid growth in the size of image databases. Despite advances in image data capture and storage techniques, development of methods for effective image retrieval has not kept pace with the technology of image production. The ability to effectively retrieve non-alphanumeric data is a complex issue. Research and development in recent years have focused on the retrieval of images by their content. In this thesis, based on Formal Concept Analysis (FCA), a new image indexing and retrieval technique is proposed. This technique allows us fast retrieval of image from the databases. The retrieval efficiency in this scheme depends on the number of attributes rather than the number of images in the database with dynamic support for addition of new images but requires an advanced knowledge of a specific domain. |
