Scribd
Upload
29 views8 pages

Hierarchical Clustering

This document discusses binarization of gray level images through hierarchical clustering algorithms. It explains agglomerative clustering, which starts with each data point as its own cluster and iteratively merges the two closest clusters. It also describes divisive clustering, which starts with all data points in one cluster and iteratively splits clusters based on dissimilarity. Finally, it discusses different linkage criteria for determining distances between clusters, including single, complete, average and centroid linkage, and notes the high computational cost of hierarchical clustering methods.

Uploaded by

ArunKumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
29 views8 pages

Hierarchical Clustering

This document discusses binarization of gray level images through hierarchical clustering algorithms. It explains agglomerative clustering, which starts with each data point as its own cluster and iteratively merges the two closest clusters. It also describes divisive clustering, which starts with all data points in one cluster and iteratively splits clusters based on dissimilarity. Finally, it discusses different linkage criteria for determining distances between clusters, including single, complete, average and centroid linkage, and notes the high computational cost of hierarchical clustering methods.

Uploaded by

ArunKumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

1

Binarization of Gray Level Images










2

How to partition the gray values into TWO clusters ?



























3
Hierarchical Clustering

Hierarchical clustering algorithm has two versions: agglomerative clustering
and divisive clustering.

Agglomerative clustering is based on the union between the two nearest
clusters. The initial condition is realized by setting every datum as a cluster.
After a few iterations it reaches the final clusters wanted. Basically, this is a
bottom-up version.

Divisive clustering starts from one cluster containing all data items. At each
step, clusters are successively split into smaller clusters according to some
dissimilarity. Basically this is a top-down version.



Agglomerative clustering :




Clustering of 8 points Dendrogram

4
In single-linkage clustering, the distance between one cluster and another
cluster is equal to the shortest distance from any member of one cluster to
any member of the other cluster:

D(C
i
,C
j
) = min {d(a,b): a in C
i
and b in C
j
}


In complete-linkage clustering, the distance between one cluster and
another cluster is equal to the greatest distance from any member of one
cluster to any member of the other cluster:

D(C
i
,C
j
) = max {d(a,b): a in C
i
and b in C
j
}


In average-linkage clustering, the distance between one cluster and
another cluster is equal to the average distance from any member of one
cluster to any member of the other cluster:

D(C
i
,C
j
) =
j i
C b
C a
C C b a d
j
i
/ )} , ( {



In centroid-linkage clustering, the distance between one cluster and
another cluster is equal to the distance between the centroids of the two
clusters:

D(C
i
,C
j
) = d(c
i
,c
j
) where c
i
and c
j
are the centroids of C
i
and C
j



High Computational Cost :

The main weakness of hierarchical clustering methods is that they do not
scale well: the time complexity is at least O(N
2
), where N is the total number
of objects.







5




(a) Original Data








(b) Agglomerative clustering

6
An example of clustering 4 data items in 2-dimensional space






7









8

You might also like