A Design of Retrieval System for Presentation Documents Using Content-Based Image Retrieval

• Published in: Advanced Communication and Networking pp. 278 - 285
• Main Authors: Lee, Hongro, Choi, Kwangnam, Choi, Kiseok, Kim, Jaesoo
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg
• Series: Communications in Computer and Information Science
• Subjects: content-based image retrieval; feature extraction; information retrieval; power-point; slidesearch engine
• ISBN: 9783642233111; 3642233112
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-642-23312-8_35

In recent year, a number of text-based slide search engines on the website for looking for power-point files have opened. However, general user wants to search information by an intuitive method which detects things based on their knowledge. In this paper, we propose an intuitive retrieval method for searching the presentation slide using content-based image retrieval. The proposed method uses the feature of digital image transformed from each slide to calculate similarity between stored slide and query slide instead of texts in the presentation slide or file name. The local feature and global feature was extracted in digital image transformed from slide and we rapidly calculate the similarity between features. We showed that similar slide can be efficiently and intuitionally searched by using content-based image retrieval technique.