You are on page 1of 16

WELCO

ME
Blurred and De-blurred of Image with Moving Average Filter in
MATLAB
PRESENTED BY

MD. RASHED RAHMAN


STUDENT MEMBER OF
PRESIDENCY UNIVERSITY

UNDER THE GUIDANCE OF

PARTHA PRATIM PAUL


ASSISTANT PROFESSOR
PRESIDENCY UNIVERSITY
CONTENTS

1. Introduction
2. Blurring
3. De-Blurring
4. The Moving Average
5. Working principles
6. Result
7. Advantages & Disadvantages
8. Conclusion
Introduction

The goal of the image restoration approach, like image


enhancement, is to enhance an image in a predetermined way. There
isn't much of a difference between image enhancement and image
restoration. Image enhancement is primarily subjective when
compared to image restoration, which is typically an objective
process. Contrast stretching, for example, is considered an
enhancement strategy because it is primarily motivated by the
appealing qualities it can display to the viewer, whereas eliminating
image blur with a de-blurring function is considered a restoration
technique.
Blurring

Blur is an un-sharp image area caused by camera or subject movement, poor focusing, or the use
of a shallow depth of field aperture. Blur effects are filters that smooth transitions and reduce
contrast by averaging pixels near defined lines' hard edges and areas with significant color
transition.

Blurring effects in digital images are classified into three types:

1. Average Blur
2. The Gaussian Blur
3. The Motion Blur
DE-BLURRING

De-blurring is the process of removing blurring artifacts from images

Techniques for De-blurring

1. The Wiener Filter


2. Filter with Regularization De-blurring
3. Lucy-Richardson Algorithm Technique
4. Deconvolution Algorithm Technique
THE MOVING AVERAGE

The moving average is the most common filter in DSP, mainly


because it is the easiest digital filter to understand and use. In spite
of its simplicity, the moving average filter is optimal for a
common task: reducing random noise while retaining a sharp step
response.
One of the very simple difference equations…

Taking the average, and as time passes, each coefficient is shifted, resulting in a kind
of moving average, hence the filter name "moving average."
Average

MATLAB code
METHODOLOGY

I have an image sample in the database that I want to detect using


MATLAB.
 
This project can blur and de-blur an image except for a specific
portion of it. So I'll use the concept of a moving average filter for
that. First and foremost, read the image and display it. Crop a
specific part before storing the variable, which is taken from the
original image. A filter finally displays the image for a moving
average filter.
Image Blurring
Image De-Blurring
Advantages

The most popular digital low-pass filter. Ideal for lowering noise while
preserving a sharp step response. Really smooth (time domain). Filter
coefficients with unity values, no MAC (multiply and accumulate)
operations are necessary. Conceptually, implementation is easy.

Disadvantages

Requires keeping track of past data from several time periods for each
predicted period. Overlooks a lot of complex relationships that the data
mentions. Does not react to changes that occur for a reason, such as
cycles and seasonal effects.
 
Conclusion

Motion or the addition of noise causes the image to become


blurry. If sounds are not added to the deteriorated image, the
direct image is de-blurred with the use of a moving average
filter. The moving Average Filter operates in the frequency
domain and seeks to reduce de-convoluted noise's influence at
frequencies with low signal-to-noise ratios. If noise is
introduced to a deteriorated image, the image will become
blurry and noisy.
REFERENCES

1. M. P. Ekstrom, ed., Digital Image Processing Techniques (Academic, Orlando, Fla., 1984).
2. A. Rosenfeld, A.C.Kak.,"Digital Picture Processing", (Academic, New York, 1982),Vol. 2.
3. H. Lim, K. C. Tan, B. T. G. Tan, ‘‘Edge errors in inverse and Wiener filter restorations of motion-blurred images and their
windowing treatment,’’ CVGIP: Graph. Models Image Process. 53, PP.186–195 (1991).
4. M.K.Singh, U.S.Tiwary, H.Kim,"adaptively Accelerated Lucy Richardson method for image deblur",
5. EURASIP Journal on advances in Signal
6. R.C. Gonazalez, R.E. woods, “ Digital Image Processing” , 2nd Edition, Pearson Education, Prentice Hall, 2002.
7. S.S.Al-Amri, N.V. Kalyankar, Khamtikar S.D., “Deblured Gaussian Blurred Image”, Journal Of Computing, Vol II, Issue
IV, 2010, pp. 33-35.
8. J. Wang, G. Wang, M. Jinag, “ Blind deblurring of spiral CT images based on ENR and Wiener Filter”, Journal of X-ray
science and Technology, 2005, pp. 49-60.
9. N.D. Narvekar, L.J. Karam, “An Iterative Deblurring algorithm based on the concept of just noticeable blur”.
10. S.S.Al-Amri, “Restoration and Deblured Motion Blurred Images”, International Journal of Multimedia , vol. 1, Issue I,
2011, pp.3-6.
Any Question?
Thank You!

You might also like