The Implementation of Cluster Identification of Data

Muhamad Zulkapley, Zafrrah (2007) The Implementation of Cluster Identification of Data. [Final Year Project] (Unpublished)

[thumbnail of 2007 Bachelor - The Implementation Of Cluster Identification Of Data.pdf] PDF
2007 Bachelor - The Implementation Of Cluster Identification Of Data.pdf

Download (843kB)

Abstract

In this paper, we combine few transformations to create a single unique transformation
based on Threshold, Edge Detector, Simple Skeleton and Hough Line Transformation
with information-theoretic-based criteria for unsupervised hierarchical image-set
clustering. The continuous image modeling is based on mixture of Gaussian densities.
The unsupervised image-set clustering is based on a generalized version of a recently
introduced information-theoretic principle, the information bottleneck principle. Images
are clustered such that the mutual information between the clusters and the image
content is maximally preserved. Experimental results demonstrate the pattern of the
image skeleton. Information theoretic tools are used to evaluate cluster quality.
Particular emphasis is placed on the application of the clustering for efficient image
search and verification. The application is very suit to offer authentic-looking
counterfeit checks.

Item Type: Final Year Project
Subjects: Z Bibliography. Library Science. Information Resources > ZA Information resources
Departments / MOR / COE: Sciences and Information Technology > Computer and Information Sciences
Depositing User: Users 2053 not found.
Date Deposited: 24 Oct 2013 09:46
Last Modified: 25 Jan 2017 09:45
URI: http://utpedia.utp.edu.my/id/eprint/9558

Actions (login required)

View Item
View Item