1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Palmprint recognition using bag of words model of discriminite scale invariant faeture transform method
Authors:	Hussein Zahraa Jabbar (P57677)
Supervisor:	Md. Jan Nordin, Assoc. Prof. Dr.
Keywords:	Palmprint recognition Bag of Words (BoW) Biometric identification.
Issue Date:	25-Feb-2013
Description:	A biometric system is one which requires the recognition of a pattern whereby it enables the differentiation of features from one individual to another. This research emphasizes palmprint recognition which provides a wide deployment range of authentication methods.The previous researchers have difficulty to extract the features of a palmprint, because it is effects of orientation, illumination, and scaling changes, and the accuracy rate of time performance need to be improved. However, he objective of this research to improve the accuracy performance and rate of time of the palmprint recognition by using Bag of Words (BoW) model, based on the Scale Invariant Feature Transform (SIFT) method and K-Means clustering.Currently, the proposed model in this research uses the SIFT for extraction of the interesting features of the palmprint images, and then uses clustering algorithm of K-Means for generating and clustering the codebooks, and using the code frequency histogram for representing images, and then classification by using kNearest Neighbours (kNN) classifier.Add to that, this model has shown a very good result without affecting different of orientation and illumination of objects, because it is associated with the use of a good description of SIFT features. Therefore, the clustering of features is considered the most important tool for the image classification task, where this task helps to reduce the testing features in the palmprint images.This system has been tested using two databases from Indian Institute of Technology Kanpur(IITK), and Chinese Academy of Sciences (CASIA), Beijing. The proposed method shows higher potential results by use the optimal Number BoW codebook clustering on the whole database, bringing significant improvement in the classification accuracy rates. The experiment shows 99.99% recognition rates in the IITK database and 99.98% in the CASIA database. It is a good approach to be implemented on high level biometric security,Sarjana
Pages:	102
Call Number:	TK7882.B56 H847 2013 3
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.