Module-III
Module III
Image segmentation: Detection of discontinuities: Point Line and Edge Detection - Edge
linking and boundary detection - Hough transform – Thresholding - Region based
segmentation: Region growing -Region splitting and merging - Use of motion in
segmentation
Morphological Image Processing: The structuring element, Basic operations on sets,
Erosion, Dilation, Opening and Closing
23-Detection of Points, Lines and Edges 7:09 PM 2
References
• Gonzalez and Woods, Digital Image Processing, Pearson
Education,3/e, (2008)
• The images in the presentation are taken from slides of Gonzales
and Woods and different internet websites
23-Detection of Points, Lines and Edges 7:09 PM 3
Image segmentation
▪ Segmentation divides an image into its constituent regions or objects
▪ The level to which subdivision is carried out depends on the problem being solved
▪ .i.e. segmentation should stop when the objects or region of interest in an application have
been detected
23-Detection of Points, Lines and Edges 7:09 PM 4
Image segmentation
▪ Most of segmentation algorithms are based on two basic properties of
intensity values
▪ Discontinuity
▪ Similarity
23-Detection of Points, Lines and Edges 7:09 PM 5
Detection of Discontinuities
▪ Points
▪ Lines
▪ Edges
23-Detection of Points, Lines and Edges 7:09 PM 6
Detection of Discontinuities
w-filter coefficients
z- pixel values
23-Detection of Points, Lines and Edges 7:09 PM 7
Detection of Points
Point detection can be done by using the mask :
23-Detection of Points, Lines and Edges 7:09 PM 8
Detection of Points
X-ray image of Result of point Result of
a turbine blade detection thresholding
23-Detection of Points, Lines and Edges 7:09 PM 9
Detection of Lines
▪ The masks will extract lines one pixel thick
and running in a particular direction
23-Detection of Points, Lines and Edges 7:09 PM 10
Detection of Lines
Binary image of a wire
bond mask
Result of
After
threshol
processing
ding
with -45°
filtering
line detector
result
23-Detection of Points, Lines and Edges 7:09 PM 11
First Derivative
Basic definition of the 1st derivative of a 1D function is :
f
f ( x 1) f ( x)
x
23-Detection of Points, Lines and Edges 7:09 PM 12
First Derivative
First derivative filter output
• Zero at constant intensities
• Non zero at the onset of a step or ramp
• Non zero along ramps
23-Detection of Points, Lines and Edges 7:09 PM 13
Second Derivative
The formula for the 2nd derivative of a function is as follows:
2 f
f ( x 1 ) f ( x 1) 2 f ( x )
x
2
Simply takes into account the values both before and after the current value
23-Detection of Points, Lines and Edges 7:09 PM 14
Second Derivative
Second derivative filter output
Zero at constant intensities
Non zero at the onset and end of a step or ramp
Zero along ramps of constant slope
23-Detection of Points, Lines and Edges 7:09 PM 15
Derivatives
• Differentiation - the rate of change of a function
• Consider a simple 1 dimensional example
23-Detection of Points, Lines and Edges 7:09 PM 16
First and Second Derivatives
A B
0 -1 -1 -1 -1 -1 0 0 6 -6 0 0 0 1 2 -2 -1 0 0 0 7 0 0 0
-1 0 0 0 0 1 0 6 -12 6 0 0 1 1 -4 1 1 0 0 7 -7 0 0
23-Detection of Points, Lines and Edges 7:09 PM 17
First and Second Derivative
• 1st derivative is constant and produces thick edges
• 2nd derivative have strong responses for fine details
• 2nd derivatives produces a double edge response at ramp and step transitions
in intensity
• The sign of the second derivative is used to determine whether an edge is a
transition from light to dark or dark to light
23-Detection of Points, Lines and Edges 7:09 PM 18
Edge Detection
An edge is a set
of connected
pixels that lie on
the boundary
between two
regions
23-Detection of Points, Lines and Edges 7:09 PM 19
Edge Detection
Roof edge
23-Detection of Points, Lines and Edges 7:09 PM 20
Edges & Derivatives
▪1st derivative tells us
where an edge is
▪2nd derivative can
be used to show
edge direction
23-Detection of Points, Lines and Edges 7:09 PM 21
Derivatives & Noise
Derivative based edge detectors are extremely sensitive to noise
23-Detection of Points, Lines and Edges 7:09 PM 22
Common Edge Detectors
▪Given a 3x3 region of an image the following edge detection filters can be used
23-Detection of Points, Lines and Edges 7:09 PM 23
Common Edge Detectors
Prewitt and Sobel Operators
for detecting diagonal edges
23-Detection of Points, Lines and Edges 7:09 PM 24
Edge Detection Example
Horizontal Gradient Component
Horizontal and Vertical
Gradient Component
Original Image
Vertical Gradient Component
23-Detection of Points, Lines and Edges 7:09 PM 25
Edge Detection Example
23-Detection of Points, Lines and Edges 7:09 PM 26
Laplacian Edge Detection
Laplacian of a 2D function is a second derivative:
Not used for edge detection as
1. It is highly sensitive to noise
2.It produces double edges and is unable to detect edge direction.
23-Detection of Points, Lines and Edges 7:09 PM 27
Laplacian Of Gaussian
Laplacian of Gaussian (or Mexican hat) filter
uses the Gaussian for noise removal and
Laplacian for edge detection
23-Detection of Points, Lines and Edges 7:09 PM 28
Laplacian of Gaussian Example
23-Detection of Points, Lines and Edges 7:09 PM 29
Edge Linking and boundary detection
23-Detection of Points, Lines and Edges 7:09 PM 30
Conclusions
▪ Detection of Points
lines
edges
23-Detection of Points, Lines and Edges 7:09 PM 31
Thank You
