UCT CS Research Document Archive

Investigating Image Processing Algorithms for Navigating Cultural Heritage Spaces using Mobile Devices

Olojede, Ayodeji and Hussein Suleman (2015) Investigating Image Processing Algorithms for Navigating Cultural Heritage Spaces using Mobile Devices. In Allen, Robert B, Jane Hunter and Marcia L. Zeng, Eds. Proceedings 17th International Conference on Asia-Pacific Digital Libraries, pages 215-224, Seoul, South Korea.

Full text available as:
PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

The use of mobile devices is increasing in the cultural heritage and museum context. The most common approach is to provide a customized mobile device to the museum visitor to navigate museum spaces. In this paper, a mobile cultural heritage guide is presented, which enables image based navigation of rock art sites using computer vision and image processing algorithms for rock art image feature detection and extraction. Traditionally such systems have used algorithms such as Scale-Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF) and Oriented Fast and Rotational Brief (ORB). The three algorithms have been integrated in a prototype and their performance has been evaluated. It was observed that digital recognition of rock art images is possible under certain image preprocessing conditions. Also, the evaluation result shows that, generally, SIFT has good accuracy and, when used in conjunction with K Nearest Neighbor, has acceptable matching speed.

EPrint Type:Conference Paper
Keywords:Cultural Heritage, Rock Art Images, Content Based Image Retrieval System
Subjects:H Information Systems: H.4 INFORMATION SYSTEMS APPLICATIONS
I Computing Methodologies: I.4 IMAGE PROCESSING AND COMPUTER VISION
H Information Systems: H.5 INFORMATION INTERFACES AND PRESENTATION
H Information Systems: H.3 INFORMATION STORAGE AND RETRIEVAL
ID Code:1063
Deposited By:Suleman, Hussein
Deposited On:26 January 2016