Scribd
Upload
7 views30 pages

Image Enhancement Mask Processing

The document discusses digital image processing techniques, focusing on image enhancement in the spatial domain, particularly through spatial filtering methods such as averaging and sharpening filters. It explains the concepts of linear and non-linear filtering, the differences between correlation and convolution, and the use of masks for processing images. Additionally, it covers advanced techniques like unsharp masking and high boost filtering to enhance image details and reduce noise.

Uploaded by

f20212675
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views30 pages

Image Enhancement Mask Processing

The document discusses digital image processing techniques, focusing on image enhancement in the spatial domain, particularly through spatial filtering methods such as averaging and sharpening filters. It explains the concepts of linear and non-linear filtering, the differences between correlation and convolution, and the use of masks for processing images. Additionally, it covers advanced techniques like unsharp masking and high boost filtering to enhance image details and reduce noise.

Uploaded by

f20212675
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 30

EEE F435 (2024-25-II)

Digital Image Processing


(Image Enhancement in the Spatial Domain :
BITS Pilani Mask Processing)
K K Birla Goa Campus
Ashish Chittora
Spatial Filtering
• Predefined operation on the neighborhood of a
pixel
• Result is stored at same coordinates as the
center pixel in a new (resultant) image
• Linear as well as non-linear filtering is possible
Linear Spatial filtering
• Processed or modified value of pixel
Correlation v/s Convolution
Example: 1-dimensional
Example: 2D-Correlation
Example: 2D-Convolution
Mask processing
Mask processing
Averaging Filter
•Also called Smoothing or Low
pass filer
•Reduces the sharp transitions
•Reduces “irrelevant” detail
(smaller than mask size)
•Noise reduction
•Smoothing of false contours
•Undesired effect: edge blurring
Spatial filtering
1 1 1

1 441 301 10 9 103 Original image

1 231 321 11 52 4

50 17 2 86 21 Resultant image

41 6 87 3 23 14 17 16 21 19

23 12 16 4 9

1 1 1

1/9 X 1 1 1

1 1 1
Averaging filter
Averaging Filter
Averaging Filter

• Wieghted Averaging

• Example
Sharpening Spatial Filters
• objective of sharpening is to highlight fine
detail in an image or to enhance detail that
has been blurred
• Averaging filter (analogous to integration)
produces blurring effect
• So sharpening must be a differentiation
operation
• Enhances intensity discontinuities
• Deemphasizes areas with slowly varying
intensity levels
Sharpening Spatial Filters

Intensities along the


Gray image (with 3 bits/pixel or centre line of image
8 intensity levels)
Sharpening Spatial Filter
Sharpening Spatial Filter
• First order derivative:

• Second order derivative:


Sharpening Spatial Filter : second derivative

• Second order derivative for enhancement:


– Laplacian operator

– For x-direction

– For y-direction

– Finally we get
Sharpening Spatial Filter
• This equation can be written as a 3X3 mask

» OR
Sharpening Spatial Filter
• If the Diagonal neighbors are also included

» OR
Sharpening Spatial Filter
• Filter for combined effect of Laplacian filtering
and adding/subtracting
Sharpening Spatial Filter
• Filter for combined effect of Laplacian filtering
and adding

• If diagonal neighbors are also considered


Sharpening spatial filters : example
Unsharp masking
• Used in publishing industry and dark room
photography for sharpening
• Subtract blurred image form original to get the
mask image
• Add (weighted) mask to the original image

• When k = 1 , it is unsharp masking. If k>1, high


boost filtering.
High boost filtering

• Slight further generalization of unsharp


masking

»a
High boost filtering
• If we use Laplacian operator to get sharpened
image then

• Mask for high boost filtering

– OR
High boost filtering
First derivative for enhancement
• First derivative or gradient

• Magnitude

• Approximation
Gradient operator
Horizental edges vertical edges
• Sobel’s

• Robert’s
Sobel’s operator : example

You might also like