Prepared By:-

MEHUL PATEL (120010741013)

RUSHIN SHAH (120010741006)

DESIGN & IMPLEMENTATION OF MARKER CONTROLLED SEGMENTATION TECHNIQUE FOR MEDICAL

APPLICATION

1

AGENDA

Segmentation

Requirements in medical application

Introduction

Watershed transformation

Over segmentation

Solutions

Marker controlled watershed transform

Morphological reconstruction

Marker & Mask

Algorithm & Results

Conclusion

References

2

WHAT IS SEGMENTATION ? ? ?

Segmentation is the process of partitioning a digital

Image into multiple segments/regions. The main aim of

segmentation is to extract the ROI (Region Of Interest)

for image analysis.

Several methods and approaches are introduced into

the area of segmentation among them a well known

method is watershed algorithm.

3

REQUIREMENT OF

SEGMENTATION IN MEDICAL

APPLICATIONS Medical image segmentation is a very important field

for the medical science. In medical images, edge

detection is an important work for object recognition of

the human organs such as brain, heart or kidney etc.

and it is an essential pre-processing step in medical

image segmentation.

Medical images such as CT, MRI or X-Ray visualizes

the various information’s of internal organs which is very

important for doctors diagnoses as well as medical

teaching, learning and research.

It is a tough job to locate the internal organs if images

contains noise or rough structure of human body

organs.4

INTRODUCTION

Image segmentation is play a vital role in most medical

image analysis task.

There are two main approaches to segmentation:

1.The frontier approach

2.The region approach

The segmentation by watershed combine the two

approaches

The major problem of the watershed transform is over-

segmentation.

5

INTRODUCTION (CONT..)

Indeed, this algorithm is sensitive to any local

minimum in the image, and tends to define the lines of

the watershed transform where each local minimum

gives rise to a region.

To avoid this problem, powerful tools adapted to

different problems have been proposed:-

1. Either reduce the number of minima and avoid

calculation of too many regions.

2. Proceed by either filtering techniques by merging

the regions according to similarity criteria after

spectral and spatial application of the watershed.

We have used ‘MARKERS’ to reduce the number of

regional minima.6

WATERSHED TRANSFORM

The watershed segmentation technique has been widely

used in medical image segmentation. Watershed

transform is used to segment gray matter, white matter

and cerebrospinal fluid from Magnetic Resonance (MR)

brain images.

There are two main approaches to segmentation: the

frontier approach and the region approach.

The segmentation by watershed combine these two

approaches & this is a powerful technique for rapid

detection of both edges and regions.

The watershed transform is a morphological gradient-

based segmentation technique.

7

The method originated from mathematical morphology

that deals with the topographic representation of an

image.

Watersheds are one of the typical regions in the field of

topography.

A drop of the water falling it flows down until it reaches

the bottom of the region.

Monochrome image is considered to be a height surface

in which high-altitude pixels correspond to ridges and

low altitude pixels correspond to valleys.

This suggestion says if we have a minima point, by

falling water, region and the frontier can be achieved.

Watershed uses image gradient to initial point and

region that can get by region growing.

WATERSHED TRANSFORM

(CONT..)

8

Here the given image lets considered as topographic

edge.

Watersheds are define as the lines separating

catchment basins, which belongs to different minima.

WATERSHED TRANSFORM

(CONT..)

If one combines the

grey level of each

point at an altitude

then it is possible to

define the

watershed

transform as

the ridge forming

the boundary

between two

watersheds.

9

A watershed can be imaged as a high mountain that

separates two region.

Each region has its own minimum and if a drop of water

falls on the one side of the watershed, it will reach

minimum of the regions.

The regions that the watershed separates are called

catchment basins.

This is to compute the

watershed of the said relief. Watersheds thus obtained

correspond to regions of the image.

Watershed represents the boundaries between adjacent

catchments. The minimum can be

interpreted as markers of watershed regions and the

watershed can be interpreted as contours,

WATERSHED TRANSFORM

(CONT..)

10

Over - Segmentation

The image will be over segmented if there are many

more minima in the image than the objects of interest.

The use of the single watershed algorithm does not

really allow good segmentation because far too many

regions are detected which causes over segmentation

and sensitivity to false edges.

Original image Watershed regions11

There are two main methods to limit this over

segmentation:

1. Hierarchical watershed segmentation.

2. Watershed by markers

The hierarchical segmentation: The hierarchical

approach is to generate a tree of regions from the

result of the watershed. Regions and watershed are

first indexed, and then the process of hierarchical

segmentation process merges the regions whose

borders are the lowest. The result is a tree in which it

is possible to explore the different levels of fusion

regions.

Over – Segmentation (CONT..)

12

Marker Controlled Watershed

Transform

We have used markers to reduce the number of

regional minima. The concept of markers is a good

approach to control over segmentation.

The markers are connected component of an image.

There are internal markers and external markers where

internal markers are associated with object of interest

and external markers are associated with the

background.

Watershed by markers based on morphological

reconstruction.

13

MORPHOLOGICAL

RECONSTRUCTION Morphological reconstruction can be thought of

conceptually as repeated dilations of an image, called

the marker image, until the contour of the marker image

fits under a second image, called the mask

image(original).

In morphological reconstruction, the peaks in the marker

image "spread out," or dilate.

Unique properties:

1. Processing is based on two images, a marker and a

mask, rather than one image and a structuring element.

2. Processing is based on the concept of connectivity,

rather than a structuring element.

3. Processing repeats until stability; i.e., the image no

longer changes.14

Repeated Dilations of Marker Image, Constrained by

Mask

Each successive

dilation is

constrained to lie

underneath the

mask.

When further

dilation ceases to

change the

image,processing

stops.

The final dilation is

the reconstructed

image.

MORPHOLOGICAL

RECONST.(CONT..)

15

MARKER AND MASK

Morphological reconstruction processes one image,

called the marker, based on the characteristics of

another image, called the mask.

The high points, or peaks, in the marker image specify

where processing begins. The processing continues

until the image values stop changing.• Perform these steps:

1. Create a marker

image.

2. Call the imreconstruct

function to

morphologically

reconstruct the image

16

MARKER AND MASK (CONT..)

marker = imsubtract(A,2)recon = imreconstruct(marker,

mask)

The choice of the constant h is very important, it depends

on the processed image, and we must choose the right

value of this constant.17

ALGORITHM

Now we see the segmentation method based on

morphological operation, and then watershed applied to

grey level images.

The key points of this method are:

1. Morphological reconstruction

2. Extract the markers of regions

3. Application of watershed transform

The watershed is the separation lines of greatest

intensity in an image.

The markers will define the sources from which the

algorithm of the watershed will simulate.

18

Steps of the method.

1. Read the original image I.

2. Morphological reconstruction of the I.

3. To detect the minimum, compute the complement of

image obtained by the morphological reconstruction, the

result image noted Ic.

4. For markers of the original image, subtract from the

original image I, the image Ic:

Mr = difference = I – Ic or Mr = I - h

5. Extended and imposed minimum, we obtained the

markers.

6. Compute the watershed transform of the markers

7. Show the watershed segmented image.

ALGORITHM(CONT..)

19

RESULTS

As per method we have considered different, grey level medical images ( cell, muscle, brain, foot and dental image).

Now, We compared the results obtained by watershedwith markers with those images that obtained bywatershed without markers.

Figure show the results of segmentation.

a. Initial image a,

b. Watershed applied to the image a,

c. Image markers

d. Watershed applied to the image c

e. Superposition of the initial image with watershed.

20

21

22

OBSERVATION

The results of segmentation with marker watershed

transforms shows the clarity and detection of objects

(region and contour) marked by the image markers c.

The result obtained by watershed without markers

(Figure b) gives no information on the regions of the

original image.

Against the result obtained by watershed with markers

shows the speed of segmentation and no more than 2

seconds in all tests.

Thus this method detects all objects of the original

image (Figure d and e).

23

OBSERVATION (CONT..) In table, the number of regions obtained was decreased; the

value of the constant h used in the reconstruction.

The problem of local minima is eliminated here. Hence,the problem of over-segmentation is solved.

Finally a simple algorithm we shown, that is fast, easy toimplement and get over the problem ofoversegmentation. The algorithm is useful to segmentthe objects that touch each other in an image.

Image I1 I2 I3 I4 I5 I6

Value of

constant

14 5 3 30 47 45

Number of

regions without

markers

2772 715 3253 1047 16450

3

8376

Number of

regions with

markers

107 72 143 31 95 51

24

MODIFIED ALGORITHM The marker controlled watershed transform is mainly for

the problems where adjacent objects are there in an

image and we have to separate them using image

processing operations.

In the initial step 1st takes the gray scale image and

compute the gradient magnitude as the segmentation

function where gradient is highest at the borders of the

object and generally low inside the object.

Now uses the internal marker to distinguish the

foreground of adjacent objects. The background of the

image will then be segregated from the foreground

objects using the external markers.

Finally aggregate the computed result of the watershed

transform and study the final image.25

Original image Gradient magnitude from

grayscale image

Estimation of the segmentation function Foreground objects by estimating internal

markers

Background objects by estimating external markers Object boundaries by applying

watershed transform

CONCLUSION

Segmentation watershed by markers based on

morphological operation is able to segment real

medicals images.

We have calculated automatically region markers. The

markers are used to control the watershed to obtain

good results.

The markers are generated automatically positioned

and thereby avoiding the problem of oversegmentation,

and manual marking.

The results show the good performance of this

approach. This approach may be used in the proper

detection of the region of interest & for problems of

decision support in medical diagnosis.27

REFERENCE

The International Journal of Multimedia & Its

Applications (IJMA) Vol.4, No.3, June 2012

International Journal of Emerging Technology and

Advanced Engineering Website: www.ijetae.com

(ISSN 2250-2459, Volume 2, Issue 4, April 2012)

MATLAB 7/Image processing toolbox/Demos/Image

Segmentation/Marker-Controlled Watershed

Segmentation

Book:- Digital Image Processing by Rafael C.

Gonzalez & Richard E. Wood [Morphological

Reconstruction(656-679)].

28

29