Scribd
Upload
121 views24 pages

Segmentation by Morphological Watersheds The Use of Motion in Segmentation

The document discusses image segmentation techniques including morphological watershed segmentation and using motion cues. [1] Morphological watershed segmentation works by treating an image as a topographical map and finding watershed ridge lines to separate regions. In MATLAB, the watershed transform computes a label matrix identifying watershed regions. [2] Accumulative difference imaging (ADI) can be used to extract motion components from image sequences by considering changes over multiple frames to ignore noise. The positive ADI shows locations of moving objects while the absolute and negative ADIs indicate motion direction and speed.

Uploaded by

Dhurgham Al-Hamdani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
121 views24 pages

Segmentation by Morphological Watersheds The Use of Motion in Segmentation

The document discusses image segmentation techniques including morphological watershed segmentation and using motion cues. [1] Morphological watershed segmentation works by treating an image as a topographical map and finding watershed ridge lines to separate regions. In MATLAB, the watershed transform computes a label matrix identifying watershed regions. [2] Accumulative difference imaging (ADI) can be used to extract motion components from image sequences by considering changes over multiple frames to ignore noise. The positive ADI shows locations of moving objects while the absolute and negative ADIs indicate motion direction and speed.

Uploaded by

Dhurgham Al-Hamdani
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

Advanced Image Processing

Hind Rustum Mohammed Shaban


Professor
Computer Department / Faculty of Computer Sciences and
Mathematics /Kufa University/ Iraq

Segmentation by Morphological
Watersheds
&
The use of motion in segmentation

Mohmmed haider

1
Segmentation by Morphological Watersheds

INTRODUCTION

The Watershed transformation is


a powerful tool for image
segmentation, it uses the
region-based approach and
searches for pixel and region
.similarities
The watershed concept was first
applied by Beucher and Lantuejoul
at 1979, they used it to segment
images of bubbles and SEM
metallographic pictures

2
Basic Concept
Visualize the image in 3D topography (terrain) •
spatial coordinates + graylevel 2–
:types of points 3 •
Points in a regional )a(–
minimum
Points on hills )b(–
Points on watershed )c(–
lines
?How to differentiate (b) & (c) •

3
4
Illustration
Objective: finding the watershed lines •
Construction of dam to prevent catchment basins from merging –

5
Illustration (cont’d)
Punch a hole in each regional minimal for •
water to fill up the catchment basin

6
7
In MATLAB
The IPT function watershed implements the watershed transform. It takes
an input image and (optionally) a connectivity criterion (4- or 8-
connectivity) as input
parameters and produces a labeled matrix (of the same size as the input
image) as a result. Elements labeled 1 and higher belong to a unique
watershed region, identified by their number, whereas elements labeled 0
.do not belong to any watershed region

8
Example
Watershed Transform of Binary Image
A: Original image B: Negative of image A C: Distance transform of B
D: watershed ridge lines E: result of segmentation

9
Morphological Watersheds
program algorithm

start -1
.read image -2
Remove background from the input image by morphological -3
opening ( The morphological open operation is an erosion
followed by a dilation and using the same structuring
. element for both operations )
maps the intensity values in grayscale image -4
computes a global threshold of the grayscale image -5
convert the intensity image into binary image -6
finding the complement of the binary image calculated by -7
step 6
10
finding the Distance transform ( The distance transform of a -8
binary image is the distance from every pixel to the
. ) nearest nonzero – valued pixel
Calculate the watershed transform using the IPT function -9
watershed ( that computes a label matrix identifying the
. watershed regions of the input matrix )
maps the colors to the result image from step 9 -10
show the result segmentation image -11
end -12

11
Aside

Watershed algorithm is often applied to the gradient •


.of an image, rather than to the image itself
Regional minima of catchment basins correlate nicely with–
the small value of the gradient corresponding the objects of
interest
.Boundaries are highlighted as the watershed lines–
The important property is that the watershed lines •
form a connected path, thus giving continuous
.boundaries between regions

12
In MATLAB
The IPT function ordfilt2 implements the
gradients of image

13
The Use of Motion in
Segmentation
Motion is a powerful cue used by humans •
and many animals to extract objects of
interest from a background of irrelevant
.detail
Motion arises from a relative displacement •
between the sensing system and the scene
.being viewed
Robotic applications–
Autonomous navigation–
Dynamic scene analysis– 14
15
Accumulative difference image (ADI)
?Why use ADI •
Isolated entries resulting from noise is not an insignificant •
problem when trying to extract motion components from a
sequence of images. Although the number of these entries can
be reduced by a thresholded connectivity analysis, this filtering
process can also remove small or slow-moving objects. One
way to address this problem is by considering changes at a
pixel location over several frames, thus introducing a
memory”into the process. The idea is to ignore changes that“
occur only sporadically over a frame sequence and can
therefore be attributed to random noise
16
Accumulative difference image ( cont’d)
:Three types of ADI •

17
Example

5
3

18
Absolute
ADI

19
Positive ADI

20
Negative
ADI

21
Accumulative difference image (cont’d)
.The nonzero area of the positive ADI is equal to the size of the moving object )1(
The location of the positive ADI corresponds to the location of the moving )2(
.object in the reference frame
The number of counts in the positive ADI stops increasing when the moving )3(
object is displaced completely with respect to the same object in the reference
.frame
.The absolute ADI contains the regions of the positive and negative ADI )4(
The direction and speed of the moving object can be determined from the )5(
.entries in the absolute and negative ADIs

22
Establishing a reference image
:This necessity applies particularly to situations •
Describing busy scenes –
In cases where frequent updating is required –
Object displacement can be established by monitoring •
.the changes in the positive ADI

23
24

You might also like