Enhanced image annotation using visual content and open knowledge base

Juzlinda Mohd Ghazali (P54323)

Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/499917

Full metadata record

DC Field	Value	Language
dc.contributor.advisor	Shahrul Azman Mohd Noah, Prof. Dr.	-
dc.contributor.author	Juzlinda Mohd Ghazali (P54323)	-
dc.date.accessioned	2023-10-13T09:36:10Z	-
dc.date.available	2023-10-13T09:36:10Z	-
dc.date.issued	2017-05-29	-
dc.identifier.other	ukmvital:85986	-
dc.identifier.uri	https://ptsldigital.ukm.my/jspui/handle/123456789/499917	-
dc.description	The advancement of Web applications has led to a significant growth of multimedia content sharing and accessibility. One of the most used media is images. Images on the Web are mainly reached by text-based searching where text query is mapped to text description of images. Textual description that is assigned to images is called image annotation. Although there are interactive applications that offer users' participation in describing images, often there are still conditions where images are left with crude or no description. Consequently, images with crude description will provide inadequate information whilst images with no description will be inaccessible by text-based search. Therefore, a good image annotation scheme is highly required. It is somewhat more challenging annotating images with no initial annotation. This is the case where this research caters for. Besides that, there is the semantic gap problem, that is, the lack of coincidence of the same visual data upon information that can be extracted and the interpretation derived from a user's point of view. The aim of this research is to propose a novel approach to image annotation by combining image low-level features and semantics available in open knowledge base. A unique scheme in improving image annotation was formed, extending ideas from various fields comprising computer vision and text processing together with semantic web technologies. Two main aspects of the approach are those involving image processing and those involving text processing. One of the steps in image annotation is image classification. The performance of various machine learning algorithms was compared in a comprehensive experiment conducted to determine the best classifier. Assessment is done by Receiver Operating Characteristics (ROC) analysis. Using feature extraction, initial tag population were generated by retrieving tags from the most similar images identified. The best parameters were determined by carrying out experiments to conclude the best performance produced. Finally, tags related to domain of interest were given semantic meaning by optimizing ontologies and the open knowledge base. The aim of evaluating the approach is to compare image annotation performance before and after linking to the open knowledge base. Evaluation is based on the standard performance metrics; precision, recall, and F-Measure. The thesis demonstrates that representing the identified concept of image annotation semantically is most useful in increasing image annotation performance.,Certification of Master's/Doctoral Thesis" is not available	-
dc.language.iso	eng	-
dc.publisher	UKM, Bangi	-
dc.relation	Faculty of Science and Technology / Fakulti Sains dan Teknologi	-
dc.rights	UKM	-
dc.subject	Image annotation	-
dc.subject	Semantic	-
dc.subject	Open knowledge base	-
dc.subject	Visual content	-
dc.subject	Classifier	-
dc.subject	Dissertations, Academic -- Malaysia	-
dc.title	Enhanced image annotation using visual content and open knowledge base	-
dc.type	Theses	-
dc.format.pages	131	-
dc.identifier.barcode	002675(2017)	-
Appears in Collections:	Faculty of Science and Technology / Fakulti Sains dan Teknologi

Files in This Item:

File	Description	Size	Format
ukmvital_85986+SOURCE1+SOURCE1.0.PDF Restricted Access		436 kB	Adobe PDF	View/Open

Show simple item record Recommend this item