IT472

Digital Image Processing

Asim Banerjee
Faculty Block #1, Room No. 1109
Extn. # 554
Edge - Types

IT472 - Digital Image Processing 2

Edge – Derivatives of Intensity Profiles

IT472 - Digital Image Processing 3

Gradient Operators

IT472 - Digital Image Processing 4

Gradient Operators

IT472 - Digital Image Processing 5

Gradient Operators

IT472 - Digital Image Processing 6

Gradient Operators - Application

IT472 - Digital Image Processing 7

Gradient Operators - Application

(a) Original image

a b
(b) Sobel gradient
c d
(c) Spatial Gaussian smoothing
e f g function
(d) Laplacian mask
(e) LoG
(f) Thresholded LoG
(g) Zero crossings.

IT472 - Digital Image Processing 8

 Edge Detection Steps
• There are two basic steps of
edge detection
1.Detection of short linear edge
segments (edgels)
2.Aggregation of edgels into
extended edges (maybe
parametric description)

IT472 - Digital Image Processing 9

Any Questions?

IT472 - Digital Image Processing 10

 Edgel Detection Methods

• Difference operators

• Parametric-model matchers

IT472 - Digital Image Processing 11

 Edge is Where Change Occurs

• Change is measured by derivative in

1D
• Biggest change, derivative has
maximum magnitude
• Or 2nd derivative is zero.

IT472 - Digital Image Processing 12

 Image Gradient (1/2)

• The gradient of an image:

• The gradient points in the direction of most

rapid change in intensity
• The gradient direction is given by:

– how does this relate to the direction of the edge?

IT472 - Digital Image Processing 13

 Image Gradient (2/2)

• The edge strength is given by the gradient

magnitude

IT472 - Digital Image Processing 14

 The Discrete Gradient
• How can we differentiate a digital
image f[x,y]?
– Option 1: reconstruct a continuous
image, then take gradient
– Option 2: take discrete derivative (finite
difference)

• How would you implement this as a

cross-correlation?
IT472 - Digital Image Processing 15
 The Sobel Operator (1/2)

• Better approximations of the

derivatives exist
– The Sobel operators below are very
commonly used

-1 0 1 1 2 1
-2 0 2 0 0 0
-1 0 1 -1 -2 -1

IT472 - Digital Image Processing 16

 The Sobel Operator (2/2)

–The standard definition of the Sobel

operator omits the 1/8 term
• Doesn’t make a difference for edge
detection
• However, the 1/8 term is needed to
get the right gradient value,

IT472 - Digital Image Processing 17

 Gradient Operators

(a) Roberts’ cross operator (b) 3x3 Prewitt operator

(c) Sobel operator (d) 4x4 Prewitt operator
IT472 - Digital Image Processing 18
Any Questions?

IT472 - Digital Image Processing 19

 Effects Of Noise (1/2)

• Consider a single row or column of the image

– Plotting intensity as a function of position gives a signal

• Where is
the edge?

IT472 - Digital Image Processing 20

 Effects Of Noise (2/2)

(a) (b) (c)

(a) Original image and its edge output
(b) Noise added (zero mean and σ = 0.03) and edge output
(c) Noise added (zero mean and σ = 0.09) and edge output

IT472 - Digital Image Processing 21

 Solution: Smooth First

Where is the
edge?

Look for peaks

in IT472 - Digital Image Processing 22
Derivative Theorem Of Convolution
• This saves us one operation:

IT472 - Digital Image Processing 23

 Laplacian of Gaussian (LoG)
• Consider

Where is the Laplacian of Gaussian

operator
edge?

Zero-crossings
of bottom graph
IT472 - Digital Image Processing 24
 2D edge detection filters

Laplacian of Gaussian

Gaussian derivative of Gaussian

• is the Laplacian operator:

IT472 - Digital Image Processing 25

Optimal Edge Detection: Canny (1/2)
• Assume:
– Linear filtering
– Additive Gaussian noise
• Edge detector should have:
– Good Detection. Filter responds to edge,
not noise.
– Good Localization: detected edge near
true edge.
– Single Response: one per edge.
IT472 - Digital Image Processing 26
Optimal Edge Detection: Canny (2/2)

• Optimal Detector is approximately

Derivative of Gaussian.
• Detection/Localization trade-off
– More smoothing improves detection
– And hurts localization.
• This is what you might guess from
above
(detect change) + (remove noise)
IT472 - Digital Image Processing 27
 Edge Detection Tutorials
• An interesting tutorial on edge detection
is available at
http://www.pages.drexel.edu/~weg22/edge.html
• An interesting tutorial on Canny edge
detector is available at
http://www.pages.drexel.edu/~weg22/can_tut.html

IT472 - Digital Image Processing 28

The Canny Edge Detector (1/4)

• original image (Lena)

IT472 - Digital Image Processing 29
The Canny Edge Detector (2/4)

• norm of the gradient

IT472 - Digital Image Processing 30
The Canny Edge Detector (3/4)

• thresholding
IT472 - Digital Image Processing 31
The Canny Edge Detector (4/4)

• thinning
• (non-maximum suppression)
IT472 - Digital Image Processing 32
 Non-maximum Suppression

• Check if pixel is local maximum along

gradient direction
– requires checking interpolated pixels p and r
IT472 - Digital Image Processing 33
 Predicting
the next
edge point
Assume the
marked point is an
edge point. Then
we construct the
tangent to the edge
curve (which is
normal to the
gradient at that
point) and use this
to predict the next
points (here either
r or s).

(Forsyth & Ponce)

IT472 - Digital Image Processing 34
 Hysteresis
• Check that maximum value of
gradient value is sufficiently large
– drop-outs? use hysteresis
• use a high threshold to start edge
curves and a low threshold to
continue them.

IT472 - Digital Image Processing 35

Effect of  (Gaussian kernel size)

original Canny with Canny with

• The choice of depends on desired behavior

– large detects large scale edges
– small detects fine features
IT472 - Digital Image Processing 36
 Scale Source (Forsyth & Ponce)

• Smoothing
• Eliminates noise edges.
• Makes edges smoother.
• Removes fine detail.
IT472 - Digital Image Processing 37
 Scale Space (1/5)

IT472 - Digital Image Processing 38

 Scale Space (2/5)

fine scale
high
threshold

IT472 - Digital Image Processing 39

 Scale Space (3/5)

coarse
scale,
high
threshold

IT472 - Digital Image Processing 40

 Scale Space (4/5)

coarse
scale
low
threshold

IT472 - Digital Image Processing 41

 Scale Space (5/5)
first derivative peaks

larger

Gaussian filtered signal

• Properties of scale space (with Gaussian smoothing)

– edge position may shift with increasing scale ()
– two edges may merge with increasing scale
– an edge may not split into two with increasing scale
IT472 - Digital Image Processing 42
 Gaussian - Image filter

Gaussian delta function

Laplacian of Gaussian
IT472 - Digital Image Processing 43
Any Questions?

IT472 - Digital Image Processing 44

 That’s all for now.

We shall continue in the next class.

IT472 - Digital Image Processing 45