Detecting Edges in Images

1
Detecting Edges in Images
By Sean Szumlanski
2
Agenda

• Reasons for detecting edges
• Introduction to computer images
• Defining the edge
• Methodologies for edge detection

3
Why edge detection?

• Segmenting into areas and objects of interest
• Psychological processing of images
• Tracking moving objects.

4
Introduction to images

• Images as 2D arrays of numbers
• Color images RGB values
• Black White images intensity values
• Intensity (R G B) / 3

5
What is an edge?

• Heightened rate of change of intensity
• High magnitude of gradient

6
Edge detection process

• Get rid of noise Gaussian Filter
• Greater sigma value means smoother image
• Convolve image with mask
• Non-maximal suppression
• Hysteresis thresholding

7
The Canny operator
Gx
Gy
Gradient Vector G ltGx, Gygt Magnitude G
sqrt(Gx2 Gy2)
8
Example
Convert to Grayscale
Original Image
Image Intensity
9
Example Gx and Gy
Grayscale
Gy
Gx
10
Example Magnitude
Gy
Gx
Gradient Magnitude
11
Non-maximal suppression
Gy
Gx
Edge Candidates (by Gx/Gy ratio)
12
Hysteresis Thresholding

• HIGH value
• LOW value

Magnitude must exceed HIGH to be considered an
edge
Neighboring magnitudes must exceed LOW to be
edge-worthy
13
Thesholding applied
Gradient Magnitude
Edges
Edge Candidates (by Gx/Gy ratio)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)