Scribd
Upload
143 views22 pages

Group 3-Image Smoothing - Sayak Khan

This document discusses various methods for smoothing digital images to reduce noise. It describes spatial filters like mean, Gaussian and median filters which operate directly on pixel values. Mean and Gaussian filters are linear while median is non-linear. Frequency domain filters like ideal low-pass filters operate by removing high frequency components in the Fourier domain. Smoothing improves image quality by reducing noise at the cost of some fine detail. The document provides examples and formulas for different smoothing techniques.

Uploaded by

SAYAK KHAN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
143 views22 pages

Group 3-Image Smoothing - Sayak Khan

This document discusses various methods for smoothing digital images to reduce noise. It describes spatial filters like mean, Gaussian and median filters which operate directly on pixel values. Mean and Gaussian filters are linear while median is non-linear. Frequency domain filters like ideal low-pass filters operate by removing high frequency components in the Fourier domain. Smoothing improves image quality by reducing noise at the cost of some fine detail. The document provides examples and formulas for different smoothing techniques.

Uploaded by

SAYAK KHAN
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 22

IMAGE SMOOTHING

PRESENTED BY

Sritiman Adak (13000118035)


Sayan Mukherjee (13000118054)
Sayak Khan (13000118055)
CONTENTS

• Why Smoothing?
• Image Smoothing
• Smoothing Spatial Filters
• Linear
• Mean filter
• Gaussian filter

• Non-Linear
• Min and Max Filter
• Median filter

• Frequency Filters
• Low-pass filter
• Ideal low-pass filter
• Conclusion
• Reference
Why Smoothing?
• Digital images contains a lot of noises that reduces the quality of images
• Noise is anything in the image that are unwanted or undesired information.
• Noise can be introduced in an image through:
• Light fluctuations
• Camera Sensor
• Transmission

• So, to remove various kind of unwanted noises from a digital image, Image
Smoothing is required
Image Smoothing
• Used to reduce noise within an image
• A key technology of image enhancement.
• It is a necessary functional module in various image processing software
• This improves the quality of the image
• Two main types of Smoothing filters
• Spatial Filters
• Frequency Filters
Smoothing Spatial Filters

Spatial Filters

Linear Non-linear

Mean Gaussian Min Max Median


Linear filters

• Smoothing linear filter is the average of the pixels contained in the


neighborhood of the filter mask.
• They are of two types:
• Mean Filter
• Gaussian Filter
Mean Filter
• Mean filtering is simply to replace each pixel value in an image with the
mean or the average value of its neighboring pixels, including itself
• This is an example of 3x3 normalized box-filter
• Drawback : Reduces fine image details
Example
Original Image

Image smoothened with mean filters

3X3 5x5 9x9 11x11


Gaussian Filter
• Smoothens the image by calculating weighted averages in a filter box
• Blurs the images and removes noise along with fine details
• More weightage to the central pixel and less to the neighboring.
• The farther away the neighbor, the smaller is the weight.
• Formula:

x = distance from origin in horizontal axis


y = distance from origin in vertical axis
σ  =Standard deviation
Gaussian Smoothing Example
Non-linear
• These are order-statistics filters whose response is based on ordering
(ranking) the pixel contained In the image area encompassed by the filter,
and then replacing the value of the center pixel with the value determined
by the ranking result.
• Types:
• Min and Max Filter
• Median Filter
Min and Max Filter

• The minimum filter selects the smallest value within the pixel values and the
max filter selects the largest value within the pixel values in the box.
• MAX filter is useful for finding the brightest points in an image. Removes salt
noise
• MIN filter is useful for finding the darkest points in an image. Removes
pepper noise
• Both filters require data sort
Median Filter

• It smooths a few pixels whose value differs significantly from their


surrounding pixels, without affecting any other pixels
• Removes both salt and pepper noise
• Salt and pepper noises occur due to random bit errors in communication
channel
Example
Frequency Filter

• The input image first needs to be transferred to frequency domain using


Fourier Transform
• Smoothing is fundamentally a low pass filtering operation on the image , in
frequency domain.
• faster than the spatial filters
• Inverse Fourier transform is done to get the resultant image
Operation of the frequency filter

• F(u, v): Fourier transformed image


• H(u, v): Filter function
Low-Pass Filter
• LPFs removes high frequency components from a given signal in frequency
domain and keeps only the low frequency components.
• Helps to smoothen the image

• E.g., Ideal Low Pass Filter


Ideal Low-pass filter
• Simplest low-pass filter which ‘cuts of’ all the all the high frequency
component of the Fourier transform, that are greater than a specified
distance D0 from the origin.
• The transform function

• Here D(u, v) is the distance of point (u, v) from the origin of the frequency
plane
• D0 is known as the cutoff frequency
Continued…

Before After
Conclusion

• Image Smoothing is a crucial part of Image Enhancement.


• There are a large number of approaches by which we can smoothen image to
achieve visually acceptable images
• According to the image properties and the task we are ultimately going to
perform with the same, we can choose the particular technique that is most
suitable.
References
• Websites
• https://www.slideshare.net/hiiampallavi15/smoothing-in-digital-image-processing
• https://www.geeksforgeeks.org/frequency-domain-filters-and-its-types/
• https://www.geeksforgeeks.org/matlab-ideal-lowpass-filter-in-image-processing/#:~:text=In%20the%20field%
20of%20Image,and%20preserves%20low%2Dfrequency%20components.&text=is%20the%20transition%20poi
nt%20between,is%20termed%20as%20cutoff%20frequency
.
• https://en.wikipedia.org/wiki/Median_filter
• https://en.wikipedia.org/wiki/Gaussian_blur#:~:text=In%20image%20processing%2C%20a%20Gaussian,image
%20noise%20and%20reduce%20detail
.

• Books
• Mark S. Nixon and Alberto S. Aguado. Feature Extraction and Image Processing
Thank You

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy