What are edges in an image?
Edges are those places in an image that correspond to object boundaries.
Edge direction: perpendicular to the direction of maximum intensity change (i.e.,
edge normal)
• Edge strength: related to the local image contrast along the normal.
• Edge position: the image position at which the edge is located.
1. Edge normal: The direction of the maximum intensity variation at the edge
point.
Modeling Intensity Changes
Step edge: the image intensity abruptly changes from one value on one side .
Ramp edge: a step edge where the intensity change is not instantaneous but
occur over a finite distance.
Ridge edge: the image intensity abruptly changes value but then returns to the
starting value within some short distance.
Roof edge: a ridge edge where the intensity change is not instantaneous but occur
over a finite distance.
Main Steps in Edge Detection
(1) Smoothing: suppress as much noise as possible, without destroying true edges.
(2) Enhancement: apply differentiation to enhance the quality of edges (i.e.,
sharpening).
(3) Thresholding: determine which edge pixels should be discarded as noise and
which should be retained (i.e., threshold edge magnitude).
(4) Localization: determine the exact edge location.
Criteria for Optimal Edge Detection
(1) Good detection : Minimize the probability of false positives (i.e., spurious
edges). o Minimize the probability of false negatives (i.e., missing real edges).
(2) Good localization : Detected edges must be as close as possible to the true
edges.
(3) Single response : Minimize the number of local maxima around the true edge.
Edge Detection
If a pixel’s gray-level value is similar to those around it, there is probably not an
edge at that point.
• If a pixel’s has neighbors with widely varying gray levels, it may present an edge
point.
Edge Detection Methods
Differential operations measure the rate of change in the image brightness
function.
METHODS OF EDGE DETECTION
First Order Derivative :
Gradient Methods o Roberts Operator o Sobel Operator o Prewitt Operator
Second Order Derivative o Laplacian o Laplacian of Gaussian o Difference of
Gaussian
Optimal Edge Detection : Canny Edge Detection
Roberts Operator
• Mark edge point only • No information about edge orientation • Work best with
binary images • Primary disadvantage: o High sensitivity to noise o Few pixels are
used to approximate the gradient
Sobel Operator
Looks for edges in both horizontal and vertical directions, then combine the
information into a single metric
prewitt
Similar to the Sobel, with different mask coefficients.
Laplacian Operators
• Edge magnitude is approximated in digital images by a convolution sum.
Laplacian Operators (Cont.)
• Masks for 4 and 8 neighborhoods
Optimal Edge Detector
Low error rate , localization , response.
Canny edge detector
• This is probably the most widely used edge detector in computer vision
canny
1. Filter image with derivative of Gaussian
2. Find magnitude and orientation of gradient
3. Non-maximum suppression: o Thin multi-pixel wide “ridges” down to single
pixel width
4. Linking and thresholding (hysteresis):
o Define two thresholds: low and high
o Use the high threshold to start edge curves and the low threshold to continue
them
Edge Detector Performance
Criteria o Probability of false edges
o Probability of missing edges
o Error in estimation of edge angle
o Mean square distance of edge estimate from true edge
o Tolerance to distorted edges
Performance
• Sobel and Prewitt methods are very effectively providing good edge maps
. • Roberts and Laplacian methods are not very good as expected.
