Low level images processing

Question 1

Low Level Image Processing:
- What are the different operations in images:

Answer 1

• Addition (Arithmetic)
  
  • Substraction (Arithmetic)
    
    • Multilication (Arithmetic)
  • Division (Arithmetic)
  • AND, OR, NOT, … (Logical)

Question 2

What are 4-Neighbours or 8-Neighbours?

Answer 2

It is a operation to make use of spatial relationships between pixels and calculate distance

Question 3

What is a characteristic of the Euclidean distance?

Answer 3

-It is more accurate and more expensive to calculate, because of the sqrt

Question 4

What does Intensity level slicing means?

Answer 4

It means highlighting a specific range of intensities in an image

Question 5

Is Threshold a specific case of Intensity level slicing?

Answer 5

Yes

Question 6

What kind of threshold are there?

Answer 6

• Binary
  
  • Binary Inverted
  • Truncate (thresh if src(x,y) > thresh else src(x,y))
  • To Zero (src(x,y) if src(x,y) > thresh else 0)
  • To Zero Inverted (0 if src(x,y) > thresh else src(x,y))

Question 7

• How can be Histograms defined?

Answer 7

• dims: number of parameters
• bins: number of subdivisions in each dim
• range: limits for the values

Question 8

What two operations can be made from histograms?

Answer 8

• determination and analysis of histogram
• image improvement based on the histogram

Question 9

How are Dark subject histograms? with highs very close to the start (inverse for Light subject)

Answer 9

Its highs are very close to the start (inverse for Light subject)

Question 10

How is overexposure histograms?

Answer 10

It has loss of details in bright areas (inverse for underexposure)

Question 11

What defines Contrast Stretching?

Answer 11

• stretching the bins will create better contrast in the image

Question 12

What is Histogram Equalization?

Answer 12

• extension of contrast stretching
• reshape the image based on function T(z) to make its histogram flat and wide
  - T(z) normally is cumulative function

Question 13

How is Image Smoothing or blurring done?

Answer 13

• neighbourhood averaging
• used to remove noise or a pre-processing for edge detection
• weighted array is moved over the original image while computing the pixel as a weighted average

Question 14

How many filters are there?

Answer 14

4

Question 15

What are the types of filters?

Answer 15

• Average Filter;
• Linear Filter;
• Median filter;
• Weighted Median filter

Question 16

What defines the Average Filter?

Answer 16

• low pass filter
• approximation as smoothing operation is performed in space and not in frequency domain

Question 17

What kind of Linear Filter are there?

Answer 17

• box filter
• gaussian filter
• laplace filter

Question 18

How does the Median filter work?

Answer 18

It sorts the pixels in the filter and gets the median

Question 19

And what about the Weighted Median filter:

Answer 19

It sorts the weighted pixels in the filter and gets the median

Question 20

What kind of Geometrical Transformations are there?

Answer 20

There are 2 types:
-Linear
-Non Linear

Question 21

What types of Linear transformations are there?

Answer 21

• Affine - Translation, rotation and scaling (preserves lines and parallelism)
• Projective - Particular cases of affine (does not preserve parallelism)

Question 22

What types of Non Linear transformations are there?

Answer 22

-twirl, ripple, sphere
they are used for correction of distortion, texture mapping, morphing