Building an efficient content based image retrieval system by changing the database structure

Abstract

The amount of interest in digital images is growing because of the advent of the World Wide Web. Many different industries are making use of digital images today, including medicine, design, journalism, education, and more. Images can be stored and retrieved using a variety of techniques and methods. Nevertheless, the majority of those engines are dependent on Meta data (key-words, descriptions, tags). Although efficient, costly, and able to target keywords and tags with each image, and these engines have their shortcomings. Though it may be possible to use a filtering system to give more accurate results, there are additional issues.