cs-f441: selected topics from computer science (deep ...cs-f441: selected topics from computer...

41
CS-F441: S ELECTED TOPICS FROM COMPUTER S CIENCE (DEEP L EARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari, Assistant Professor, Department of Computer Science and Information Systems, BITS Pilani, Rajasthan-333031 INDIA Oct 25, 2019 (Campus @ BITS-Pilani July-Dec 2019)

Upload: others

Post on 25-Jun-2020

13 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

CS-F441: SELECTED TOPICS FROM COMPUTER

SCIENCE (DEEP LEARNING FOR NLP & CV)

Lecture-KT-07: Edge Detection

Dr. Kamlesh Tiwari,Assistant Professor,

Department of Computer Science and Information Systems,BITS Pilani, Rajasthan-333031 INDIA

Oct 25, 2019 (Campus @ BITS-Pilani July-Dec 2019)

Page 2: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

ImageGrayscale image can be considered as a function f : R2 → R

Being digital adds quantization and sampling.

We may apply operatorson this function

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 2 / 25

Page 3: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

ImageGrayscale image can be considered as a function f : R2 → R

Being digital adds quantization and sampling. We may apply operatorson this function

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 2 / 25

Page 4: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

ImageGrayscale image can be considered as a function f : R2 → R

Being digital adds quantization and sampling. We may apply operatorson this function

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 2 / 25

Page 5: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

Use Kernal/Mask/FilterTo replace each pixel by a linear combination of its neighbors

Cross-correlationG[i, j] =

∑ku=−k

∑kv=−k h[u, v ]f [i + u, j + v ]

G = H ⊗ F

ConvolutionG[i, j] =

∑ku=−k

∑kv=−k h[u, v ]f [i − u, j − v ]

G = H × F

Convolution is commutative and associative

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 3 / 25

Page 6: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

Use Kernal/Mask/FilterTo replace each pixel by a linear combination of its neighbors

Cross-correlationG[i, j] =

∑ku=−k

∑kv=−k h[u, v ]f [i + u, j + v ]

G = H ⊗ F

ConvolutionG[i, j] =

∑ku=−k

∑kv=−k h[u, v ]f [i − u, j − v ]

G = H × F

Convolution is commutative and associative

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 3 / 25

Page 7: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 4 / 25

Page 8: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 4 / 25

Page 9: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 4 / 25

Page 10: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 4 / 25

Page 11: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Linear filtering

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 4 / 25

Page 12: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Gaussian filter

Gσ =1

2πσ2 e− x2+y2

2σ2

Act as low-pass filter as it removes high-frequency components.Convolution with self is another Gaussian.

Convolving two times with Gaussian kernel of σ width is equivalent toConvoluting once with Gaussian kernel of

√2σ width

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 5 / 25

Page 13: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Gaussian filter

Gσ =1

2πσ2 e− x2+y2

2σ2

Act as low-pass filter as it removes high-frequency components.Convolution with self is another Gaussian.

Convolving two times with Gaussian kernel of σ width is equivalent toConvoluting once with Gaussian kernel of

√2σ width

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 5 / 25

Page 14: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Sharpening

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 6 / 25

Page 15: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Sharpening Filter

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 7 / 25

Page 16: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Edge detection

Image is converted into a set ofcurves.

Gets salient features of thesceneCompact than pixels

Edges are caused by a variety of factors

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 8 / 25

Page 17: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Edge detection

Image is converted into a set ofcurves.

Gets salient features of thesceneCompact than pixels

Edges are caused by a variety of factors

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 8 / 25

Page 18: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Characterizing edges

An edge is a place of rapid change in the image intensity function

How to differentiate a digital image (filter ?)

∂f∂x

[x , y ] = F [x + 1, y ]− F [x , y ]

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 9 / 25

Page 19: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Image gradient

The gradient of an image: Of = [ ∂f∂x ,

∂f∂y ]

Points in the direction of most rapid increase in intensity

where strength is ||Of || =√( ∂f∂x )

2 + ( ∂f∂y )

2

and direction is θ = tan−1( ∂f∂y /

∂f∂x )

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 10 / 25

Page 20: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Image gradient

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 11 / 25

Page 21: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Effects of Noise

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 12 / 25

Page 22: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Solution: smooth first

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 13 / 25

Page 23: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Associative property of convolutionDifferentiation is convolution, and convolution is associative:

ddx

(f ∗ h) = f ∗ ddx

h

This can help us to saves one operation

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 14 / 25

Page 24: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Directional derivative with Gaussian filter

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 15 / 25

Page 25: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Sobel OperatorCommon approximation of derivative of Gaussian

Edge magnitude is√

s2x + s2

y and the direction is tan−1(sy/sx)

Edge is detected by using threshold

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 16 / 25

Page 26: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Sobel OperatorCommon approximation of derivative of Gaussian

Edge magnitude is√

s2x + s2

y and the direction is tan−1(sy/sx)

Edge is detected by using threshold

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 16 / 25

Page 27: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Non-maximum supression

Check if pixel is local maximum along gradient direction

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 17 / 25

Page 28: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Non-maximum supression

Check if pixel is local maximum along gradient direction

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 17 / 25

Page 29: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Non-maximum supression

Check if pixel is local maximum along gradient direction

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 17 / 25

Page 30: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Non-maximum supression

Check if pixel is local maximum along gradient direction

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 17 / 25

Page 31: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Roberts

Roberts operator is a simple approximation to the first derivative.It makes the edge point only, not the information about the edgeorientation.It works well in binary images.The Roberts method finds edges using the Roberts approximationto the derivative.√

(I(i , j)− I(i − 1, j − 1))2 + (I(i , j − 1)− I(i − 1, j))2[0 1−1 0

] [1 00 −1

]

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 18 / 25

Page 32: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Prewit Operator

Performs smoothening (by averaging) and then finds the magnitude ofderivatives and applies threshold on them.

The mask used is as below

hx =

1 0 −11 0 −11 0 −1

hy =

1 1 10 0 0−1 −1 −1

Bigger size mask is

hx =

−3 −1 1 3−3 −1 1 3−3 −1 1 3−3 −1 1 3

hy =

3 3 3 31 1 1 1−1 −1 −1 −1−3 −3 −3 −3

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 19 / 25

Page 33: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Laplacian of Gaussian (Marr-Hildreth)It is high-pass, and non-directional filter. Uses second derivatives 0 −1 0

−1 4 −10 −1 0

−1 −1 −1−1 8 −1−1 −1 −1

Apply LoG to remove noise.

0 0 1 0 00 1 2 1 01 2 −16 2 10 1 2 1 00 0 1 0 0

1 Detect all zero crossings{+,−}, {+,−}, {+,0,−},{+,0,−}

2 Use threshold to keep onlystrong ones

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 20 / 25

Page 34: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [1/3]

Standard edge detector: Probability of detecting real edge points ismaximized while the probability of falsely detecting non-edge points isminimized.

Five separate steps:

1. Smoothing: Blurring of the image to remove noise. (by applying aGaussian filter). The kernel of a Gaussian filter with a standarddeviation of σ = 1.4 is given by

1159

2 4 5 4 24 9 12 9 45 12 15 12 54 9 12 9 42 4 5 4 2

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 21 / 25

Page 35: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [1/3]

Standard edge detector: Probability of detecting real edge points ismaximized while the probability of falsely detecting non-edge points isminimized.Five separate steps:

1. Smoothing: Blurring of the image to remove noise. (by applying aGaussian filter). The kernel of a Gaussian filter with a standarddeviation of σ = 1.4 is given by

1159

2 4 5 4 24 9 12 9 45 12 15 12 54 9 12 9 42 4 5 4 2

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 21 / 25

Page 36: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [2/3]

2. Finding gradients: The edges should be marked where thegradients of the image has large magnitudes (by applyingSobel-operator).

3. Non-maximum suppression: Only local maxima should bemarked as edges.

1 Round the gradient direction to nearest 45◦, corresponding to theuse of an 8-connected neighborhood.

2 Compare the edge strength of the current pixel with the edgestrength of the pixel in the positive and negative gradient direction.i .e. if the gradient direction is north (theta = 90◦), compare with thepixels to the north and south.

3 If the edge strength of the current pixel is largest; preserve thevalue of the edge strength. If not, suppress (i .e. remove) the value.

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 22 / 25

Page 37: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [2/3]

2. Finding gradients: The edges should be marked where thegradients of the image has large magnitudes (by applyingSobel-operator).

3. Non-maximum suppression: Only local maxima should bemarked as edges.

1 Round the gradient direction to nearest 45◦, corresponding to theuse of an 8-connected neighborhood.

2 Compare the edge strength of the current pixel with the edgestrength of the pixel in the positive and negative gradient direction.i .e. if the gradient direction is north (theta = 90◦), compare with thepixels to the north and south.

3 If the edge strength of the current pixel is largest; preserve thevalue of the edge strength. If not, suppress (i .e. remove) the value.

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 22 / 25

Page 38: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [3/3]

4. Double thresholding: Edge pixels stronger than the highthreshold are marked as strong; edge pixels weaker than the lowthreshold are suppressed and edge pixels between the twothresholds are marked as weak. Strong edges are white, whileweak edges are grey. Edges with a strength below boththresholds are suppressed

5. Edge tracking by hysteresis: Final edges are determined bysuppressing all edges that are not connected to a very certain(strong) edge. Strong edges are interpreted as certain edges, andare included in the final edge image. Weak edges are included ifand only if they are connected to strong edges.

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 23 / 25

Page 39: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector [3/3]

4. Double thresholding: Edge pixels stronger than the highthreshold are marked as strong; edge pixels weaker than the lowthreshold are suppressed and edge pixels between the twothresholds are marked as weak. Strong edges are white, whileweak edges are grey. Edges with a strength below boththresholds are suppressed

5. Edge tracking by hysteresis: Final edges are determined bysuppressing all edges that are not connected to a very certain(strong) edge. Strong edges are interpreted as certain edges, andare included in the final edge image. Weak edges are included ifand only if they are connected to strong edges.

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 23 / 25

Page 40: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Canny Edge Detector

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 24 / 25

Page 41: CS-F441: Selected Topics from Computer Science (Deep ...CS-F441: SELECTED TOPICS FROM COMPUTER SCIENCE (DEEP LEARNING FOR NLP & CV) Lecture-KT-07: Edge Detection Dr. Kamlesh Tiwari,

Thank You!

Thank you very much for your attention1!

Queries ?

1https://www.cs.cornell.edu/courses/cs6670/2011sp/lectures/lec02 filter.pdf

STCS-DL4NLP&CV (CS-F441) Campus @ BITS-Pilani Lecture-KT-07 (Oct 25, 2019) 25 / 25