pillar k means

15
A New Approach for Image Segmentation using Pillar-K-means Algorithm PRENSENTED BY SWATHI. B

Upload: swathi-b

Post on 15-Aug-2015

80 views

Category:

Education


4 download

TRANSCRIPT

Page 1: Pillar k means

A New Approach for Image Segmentation usingPillar-K-means Algorithm

PRENSENTED BYSWATHI. B

Page 2: Pillar k means

This paper presents a new approach for image segmentation by applying Pillar-K-means algorithm.

This segmentation process includes a new mechanism for clustering the elements of high-resolution images in order to improve precision and reduce computation time

The system applies K-means clustering to the image segmentation after optimized by Pillar Algorithm

This algorithm is able to optimize the K-means clustering for image segmentation in aspects of precision and computation time.

Page 3: Pillar k means

3

Segmentation

Divide the image into segments.

Each segment:

– Looks uniform

– Belongs to a single object.

– Have some uniform attributes.

– All the pixel related to it are connected.

Page 4: Pillar k means

Main approaches• Histogram-based segmentation

• Region-based segmentation

– Edge detection– Region growing– Region splitting and merging.

• Clustering– K-means– C-means– Pillar-k means

4

Page 5: Pillar k means

Practical Applications of Image segmentationMedical Applications Locate tumors and other pathologiesMeasure tissue volumesComputer guided surgeryDiagnosisTreatment planningStudy of anatomical structureLocate objects in satellite images (roads, forests, etc)Face RecognitionFinger print Recognition , etc

Page 6: Pillar k means

Idea:• Determine the number of clusters

•Find the cluster centers and point-cluster correspondences to minimize error

Problem: Exhaustive search is too expensive.Solution: We will use instead an iterative search. [Recall the ideal quantization procedure.]

Algorithm

– fix cluster centers; allocate points to closest cluster– fix allocation; compute best cluster centers

K-means

Error function =

6

Page 7: Pillar k means

AlgorithmThe K-means algorithm is an iterative technique that is used to

partition an image into K clusters.

The basic algorithm is:Pick K cluster centers, either randomly or based on some

heuristic Assign each pixel in the image to the cluster that minimizes the

variance between the pixel and the cluster center Re-compute the cluster centers by averaging all of the pixels in

the cluster Repeat steps 2 and 3 until convergence is attained (e.g. no

pixels change clusters)

Page 8: Pillar k means

Example – clustering with K-means using gray-level and color histograms(from slides by D.A. forsyth)

8

Page 9: Pillar k means

C-means

The fuzzy c-means algorithm is very similar to the k-means

algorithm:Choose a number of clusters. Assign randomly to each point coefficients for being in the clusters.

Repeat until the algorithm has converged (that is, the coefficients' change between two iterations is no more than , the given sensitivity threshold) :

Compute the centroid for each cluster, using the formula above.

For each point, compute its coefficients of being in the clusters,

Page 10: Pillar k means

The algorithm minimizes intra-cluster variance as well, but has the same problems as k-means, the minimum is a local minimum.

And the results depend on the initial choice of weights.

The expectation-maximization algorithm is a more statistically formalized method which includes some of these ideas:

partial membership in classes. It has better convergence properties and is in general preferred to fuzzy-c-means.

Page 11: Pillar k means

• Pillar K-means

• The image segmentation is important to unify contiguous colors in the

color vector space into representative colors.

• It can improve significantly performance of the information extraction,

such as color, shape, texture, and structure.

• This section describes our approach for image segmentation using our

proposed Pillar algorithm to optimize K-means clustering.

• The image segmentation system pre-proceeds three steps:

Noise removal,

Color space transformation

Dataset normalization.

Page 12: Pillar k means
Page 13: Pillar k means

Initial centroid optimization of k-means clustering for image segmentation

Page 14: Pillar k means

CONCLUSION

In this project, we have presented a new approach for image

segmentation using Pillar-K-means algorithm.

The system applies K-means clustering after optimized by Pillar Algorithm.

This algorithm is able to optimize the K-means clustering for image

segmentation in

aspects of precision and computation time.

A series of experiments involving four different color spaces with variance

constraint and execution time were conducted.

Page 15: Pillar k means

THANK YOU