Edge detection Flashcards

1
Q

What is an intensity image?

A

A matrix whose values represent the intensity of light

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is an intensity gradient?

A

A function mapping coordinates to intensity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How can you find the gradient of an intensity image in the x direction?

A

Use the mask [
[-1, 1],
[-1, 1]
]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How can you find the gradient of an intensity image in the y direction?

A

Use the mask [
[1, 1],
[-1, -1]
]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Edge direction

A

Perpendicular to the direction of maximum intensity change

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the main steps in edge detection?

A

Smoothing
Enhancement
Thresholding
Localisation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are the two ways edges are detected using derivatives?

A

Detecting the local maxima or minima of the first derivative
Detecting the zero-crossings of the second derivative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

How does hysteresis thresholding work?

A

Uses a high and a low threshold. Points above the high threshold are definitely an edge. Points above the low threshold are an edge if they are connected to an existing edge.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Why are second order edge detectors better than first order?

A

First order give too many edge points, second order only give edge points that are the local maxima.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Describe the steps of canny edge detector

A

Smooth the image using a gaussian
Compute image gradient using the sobel operator
Compute laplacian along the gradient direction at each pixel
Find zero crossings in laplacian to find the edge location

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is a gaussian also known as?

A

Bell curve

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

The laplacian is

A

The second derivative of a signal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

After taking a laplacian, what must you do to find edges?

A

Find the zero crossings

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are the advantages of the canny edge detector?

A

Produces smooth, single pixel thick edges.
Resistant to noise

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the advantage of the sobel edge detector?

A

Quick to compute

How well did you know this?
1
Not at all
2
3
4
5
Perfectly