The Implementation of Cluster Identification of Data

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.