Yangon Technological University

Department of Electronic Engineering

EcE-71017 Digital Image Processing
Chapter 10-Image Segmentation

Presented by

Ma Khin Ni Ni Htet

ME.EC-4
Outline of Presentation

 Point, Line, and Edge Detection

 Background

 Detection of Isolated Points

 Line Detection

 Edge Models

 Thresholding

 Region-Based Segmentation

 Segmentation Using Morphological Watersheds

 The Use of Motion in Segmentation

 Segmentation is one major step in the Image Processing.

 The first category, the approach is to partition an image based on abrupt changes in intensity, such as edges.

 The second category are based on partitioning an image into regions.

 Thresholding, region growing, and region splitting and merging are examples of methods in this category.

 Discontinuity-based segmentation: This category assumes that boundaries between regions and background are

sufficiently different in intensity to detect them based on local discontinuities.

 Similarity-based segmentation: This category involves partitioning an image into regions where the regions are

similar according to predefined criteria.

 Image Segmentation
Technique

Discontinuities Similarity

Edge-based Clustering-based Threashold Region-based

Point, Line, edge

Region Region
Gradient based Gaussian based Global Local Growing Splitting

Sobel Candy Edge

Rober Laplacian of
Gaussian
Prewitt
 Point Detection:

 Step for point dectction:

1. Apply Laplacian filter to image to obtain R(x, y).

2. Create binary image by threshold.

Where T is a non-negative threshold.

 Line Detection:

 A special mask is needed to detect a special type of line.

 Examples:

-Horizontal mask has high response when a line is passed through the middle row of the mask.
 Edge detection:
 Edge detection is the approach for segmenting images based on abrupt changes in intensity. It is used to
detect the boundaries or to find size or location of an object in an image.

 What is an edges?
- An edge is a set of connected pixel that lie on the boundary between two regions.
- An edge is a ‘local’ concept whereas a region boundary, owing to the way, it is defined, is a more global
idea.
 Detect the three basic types of grey-level discontinuities
 Point,
 Line,
 Edges
 Image Sharping Techniques
 Gradient based (First Order)-Produce thicker edge
 Gaussian based (Second Order)-Have a strong response of fine detail, such as thin line and isolated point, and
noise.

Can be done by running a mask through the image

 R = + + ........+
=
Two kinds of approaches to segmentation: discontinuity and similarity.

 Discontinuity

The strategy is to partition an image based on abrupt changes in intensity

 Detection of gray level discontinuities:

-Point detection

-Line detection

-Edge detection

(1) Gradient operators

(2) Gaussian based operators

Similarity

The strategy is to partition an image into regions that are similar according to a set of predefined criteria.
 Edge models:

- Step edge (Ideal edge), ramp edge (thick edge), and roof edge

- The edges in Fig. 10.8 are free of noise

(a)(b)(c)

Figure (10.8) Form left to right, models (ideal representations) of a step, a ramp, and their corresponding

intensity profiles
• An image may all the three types of edges

Figure 10.9 A image showing (zoomed) actual ramp (bottom, left), step
(top, right), and roof edge profiles. The profiles are from dark to light, in the areas
indicated by the short line segments shown in the small circles. The ramp and “step”
profiles span 9 pixels and 2 pixels, respectively. The base of the roof edge is 3 pixels.
(Original image courtesy of Dr. David R. Pickens,Vanderbilt University.)
First and Second order derivatives response at the edge:

FIGURE 10.10
(a) Two regions of
constant intensity
separated by an
ideal vertical
ramp edge.
(b) Detail near
the edge, showing
a horizontal
intensity profile,
together with its
first and second
derivatives.

1. The magnitude of the first derivative can be used to detect the presence of an edge at a point.
2. Sign of second derivative indicates which side of the edge the pixel is on.
3. Zero crossings of the second derivative can be used to locate the centers of thick edges.
 First Second
Edge Models derivative order
 All images have 8 bits of intensity resolution, with 0 and 255 representing black and white.
 There are three fundamental steps performed in edge detection:
(1) Image smoothing for noise reduction.
(2) Detection of edge points.
(3) Edge localization.

FIGURE 10.11 First column: Images and intensity profiles

of a ramp edge corrupted by random Gaussian noise of zero
mean and standard deviations of 0.0, 0.1, 1.0, and 10.0
intensity levels, respectively. Second column: First-derivative
images and intensity profiles.Third column: Second-derivative
images and intensity profiles.
• Laplace is sharp edges detection
• Gaussian is reduce the noise.

Figure 10.12 Using the gradient to determine edge strength and direction at a point.
Note that the edge is perpendicular to the direction of the gradient vector at the point
where the gradient is computed. Each square in the figure represents one pixel.
This expression is called the Laplacian of a Gaussian (LoG).

FIGURE 10.21
(a) Three dimensional plot
of the negative of
the LoG. (b) Negative of the
LoG displayed as
an image. (c) Cross section of
(a) showing zero
crossings.
(d) Mask approximation to
the shape in (a).The negative
of
this mask would be used in
practice.
More Advanced Techniques for Edge Detection
 (1)The Marr-Hildreth edge detector (Can find method of matrix)
 (2) The Canny edge detector (differential method first derivative of a Gaussian: )
(i) Low error rate
(ii) Edge points should be well localized
(iii) Single edge point response

 1. Smooth the input image with a Gaussian filter.

 2. Compute the gradient magnitude and angle images.
 3. Apply nonmaxima suppression to the gradient magnitude image.
 4. Use double thresholding and connectivity analysis to detect and link edges.
Segmentation Using Morphological Watersheds
We have discussed segmentation based on three principal concepts:
 (1) edge detection,
 (2) thresholding, and
 (3) region growing.

What is Image Segmentation: The Basics and Key Techniques

 What is Image Segmentation?
 The Various Stages of Image Segmentation
 Types of Image Segmentation
 What Image Segmentation Techniques are used to Annotate Data?
 Applications of Image Segmentation
 Object detection and classification of interior objects

What Image Segmentation Techniques are Used to Annotate Data?

 Here are some common image segmentation techniques:
 Edge-Based Segmentation. This is a popular
 shows a gray-scale image
 is a topographic view,

Figure:
(a) Original image.
(b) Topographic
view. (c)–(d) Two
stages of flooding.
Figure:
(a) Image of blobs.
(b) Image gradient.
(c) Watershed lines.
(d) Watershed lines
superimposed on
original image.
(Courtesy of Dr.
S. Beucher,
CMM/Ecole des
Mines de Paris.)
Figure: ADIs of a rectangular object moving in a southeasterly direction. (a) Absolute
ADI. (b) Positive ADI. (c) Negative ADI.
 Where to use the image segmentation?
 Image segmentation is a key building block of computer vision technologies and algorithms.
 It is used for many practical applications including medical image analysis,
 computer vision for autonomous vehicles,
 face recognition and detection,
 video surveillance, and satellite image analysis.

 Advantages of image segmentation include easy data storing, faster processing, and a multiple copy
system.
 Disadvantages include the need for extreme precision and error-prone results..
Thank you for your attention