Auto tagging on smart search internet

• Main Author: Ching, James Chuen Keong.
• Other Authors: Chen Lihui
• Format: Final Year Project
• Language: English
• Published: 2009
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/15704
• Institution: Nanyang Technological University

The use of tagging is one of the most important and popular features of the second generation web development and design (Web 2.0). The key highlight of using a tag is to give a keyword or a term to a piece of information to describe the contents of object for images. With these tags organized, users can efficiently browse and search the information they may require using a search engine. However, tagging is often very time consuming and tedious to do especially on images. With over billions of images hosted on the internet and they are neither tagged nor described. These images cannot be searched or be used to find similar information as there is no metadata bind to it. This would deem the images as useless other than displaying it where the user has posted it. As there is a strong growing number of images uploaded every day, this makes it ideal to develop a prototype automated image tagging system. This system must be able to tag the images fairly accurately and efficiently to generate the result at real time. The proposed system makes use of the color histogram for a few customized comparison methods and the result will be discussed. This report highlights a prototype developed for image tagging. The prototype system is built as a platform for research and development in image tagging.