FINAL YEAR PROJECT PRESENTATION

NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR

Topic: Image Restoration using Adaptive Filters

S.No. Name Scholar ID
1 Masina Niranjan Kumar 1913121
2 Ammu Rammurthy Kaushik 1913149
3 Sampathirao Karthik Raju 1913156

Under the Guidance of

Dr. Tapan Pradhan
Electrical Engineering department
 CONTENTS

1. What is Image Restoration?

2. Literature Review
3. Motivation
4. Objective of this Project
5. Image Degradation/Restoration Model
6. Objectives of Image Restoration
7. What is Noise?
8. Types of Noise occurring usually
9. Reduction of Noise in Spatial Domain
10. Reduction of Noise in Frequency Domain
11. Adaptive Filters
12. Work plan
13. Conclusion
 WHAT IS IMAGE RESTORATION?

Image restoration attempts to restore images that have been degraded.

• Identifies the degradation process and attempts to reverse it
• Similar to image enhancement, but more objective.

Before After

Fig.1:-Comparison of original image and image after restoration

Literature Review
Title of the Research
S.No. Publication Details Learnings
Paper/Reference Book

1 Digital Image Processing, By Rafael C. Gonzalez, University of Tennessee. Fundamentals in Digital Image
Third Edition. Richard E. Woods, Processing, and basics of noise
MedData Interactive filtering.

2 Performance Comparison M. Yadav and N. Tiwari, "Performance Focused on CNN based state of art
of Image Restoration Comparison of Image Restoration techniques denoiser methods and perform a
techniques using CNN and using CNN and their applications," 2021 5th comparative study on various methods
their applications International Conference on Computing
Methodologies and Communication (ICCMC),
2021, pp. 1146-1151, doi:
10.1109/ICCMC51019.2021.9418340.

3 A Survey of Restoration W. Zhang, L. Dong, X. Pan, P. Zou, L. Qin and This paper introduced overview of
and Enhancement for W. Xu, "A Survey of Restoration and underwater image restoration and
Underwater Images Enhancement for Underwater Images," in IEEE enhancement techniques which are
Access, vol. 7, pp. 182259-182279, 2019, doi: broadly classified as Image
10.1109/ACCESS.2019.2959560. restoration(Physical based model) and
Image enhancement(Non physical
 Title of the Research
S.No. Publication Details Learnings
Paper/Reference Book

4 Underwater Image Restoration W. Luo, S. Duan and J. Zheng, "Underwater As some details of the processed image
and Enhancement Based on a Image Restoration and Enhancement Based might disappear in histogram
Fusion Algorithm With Color on a Fusion Algorithm With Color Balance, equalization, there might be excessive
Balance, Contrast Optimization, Contrast Optimization, and Histogram enhancement at the peak of histogram.
and Histogram Stretching Stretching," in IEEE Access, vol. 9, pp. For wavelet transform, it is usually
31792-31804, 2021, doi: successful to deal with images captured
10.1109/ACCESS.2021.3060947. in shallow waters, while fails in deep
waters where the red light attenuation is
severe.
5 Application of fuzzy image Shuo Sun, Application of fuzzy image Image restoration based on conventional
restoration in criminal restoration in criminal investigation, filter algorithms cannot achieve
investigation Journal of Visual Communication and satisfactory performance. Thus, they first
Image Representation, Volume 71, introduced the image restoration
2020,102704, ISSN 1047-3203, algorithms based on different
https://doi.org/10.1016/j.jvcir.2019.102704. degradation model. Then, they proposed
some applications of fuzzy image
restoration in criminal investigation.
6 Fast video enhancement Sungjin Kim and Joonki Paik, "Fast video A fast video enhancement algorithm is
algorithm using real-time image enhancement algorithm using real-time proposed. The algorithm enhances the
restoration framework image restoration framework," Proceedings quality of video by removing
of 2002 IEEE 10th Digital Signal degradation due to subsampling and
Processing Workshop, 2002 and the 2nd compression.
 MOTIVATION

• Image restoration is the problem of recovering images which have been

degraded by blurring and noise.
• We often find ourselves in the need for a platform that restores
degraded images, be it old images or incomplete images due to
damaged sensors etc.,
• Since imaging devices are never perfect, there are many applications
for image restoration: astronomy, medical imaging, remote sensing,
microscopy, underwater imaging, near infrared imaging etc.,
 OBJECTIVE OF THIS PROJECT

Restoration of Images using adaptive filters:

• To develop an optimized approach to restore images that consists of
different types of noise and blur components.
• To compare this approach with existing traditional image restoration
methods and improve the performance in different conditions.
 IMAGE DEGRADATION/RESTORATION MODEL

• Image Restoration is the process of recovering an image that has been degraded by some knowledge of
degradation function H and the additive noise term (x, y). Thus in restoration, degradation is modelled and its
inverse process is applied to recover the original image.

Fig.2:-Image Degradation Model

 OBJECTIVES OF IMAGE RESTORATION

• The major objective of Image restoration is to obtain an estimate of the original image,

• The more the resemblance of our restored image ( ) with the original image ( ),
the greater the efficiency of the restoration process.

• Image restoration can be done in

(a) Spatial Domain
(b) Frequency Domain
 WHAT IS NOISE?

• Image noise is random variation of brightness or color information in images, and is usually an
aspect of electronic noise.
• It can be produced by the image sensor and circuitry of a scanner or digital camera.
• The principle source of noise in digital images arise during image acquisition (digitization) and
transmission.
• The performance of imaging sensors is affected by a variety of factors.
• Images are corrupted during transmission due to interference in channel.
TYPES OF NOISE OCCURRING USUALLY:

• The below mentioned noises are commonly found

1. Gaussian Noise
2. Rayleigh Noise
3. Erlang (Gamma) Noise
4. Exponential Noise
5. Uniform Noise
6. Impulse (Salt & Pepper) Noise
REDUCTION OF NOISE IN SPATIAL DOMAIN USING MEAN
FILTERS

• Types of Mean Filters are:-

1. Arithmetic Mean Filter
2. Geometric Mean Filter
3. Harmonic Mean Filter
4. Contra Harmonic Mean Filter
• Let represent the set of co-ordinates in the rectangular subimage window
of size mxn, centred at point (x,y)
REDUCTION OF NOISE IN SPATIAL DOMAIN USING ORDER-
STATISTICS FILTERS

• Order-Statistics Filters are spatial filters whose response is based on ordering(ranking) the pixels
contained in the image area encompassed by the filter

• Types of Order-Statistics Filters are:-

1. Median Filters
2. Max and Min Filters
3. Midpoint Filters
4. Alpha-trimmed Mean Filters
 REDUCTION OF NOISE IN FREQUENCY DOMAIN

Band Pass Filters

• The applications of band reject filtering is for noise removal in applications where the general location of the
noise component(s) in the frequency domain is approximately known.

Band Reject Filters

• A bandpass filter performs the opposite operation of a bandreject filter. The transfer function of a bandpass
filter is obtained from a corresponding bandreject filter with transfer function HBR(u,v) by using the equation

Notch Filters
• A notch filter rejects (or passes) frequencies in predefined neighbourhoods about a centre frequency. Due to
the symmetry of the Fourier transform, notch filters must appear in symmetric pairs about the origin in order
to obtain meaningful results.
 INVERSE FILTERING

• The simplest approach to restoration is direct inverse filtering, where we compute an estimate, of
the transform of the original image simply by dividing the transform of the degraded image, , by
the degradation function

• It tells us that even if we know the degradation function we cannot recover the undegraded image
[the inverse Fourier transform of ] exactly because is not known.
 WEINER FILTER

• This image restoration technique may be a commonplace approach that has been planned by N.
Weiner, which contains each the degradation perform and applied math representative of noise into the
restoration perform
• It’s thought of united of the simplest deblurring direct ways by that image will be reconstructed from
degraded one, by mistreatment renowned foreign terrorist organization
• It mechanisms with each great and low pass filter to execute deconvolution tom take away noise with
density operation. f = g × (f + n)
• The yield image is gotten by higher than equation wherever additive noise and frequency
physiognomies area unit renowned.
 ADAPTIVE FILTERS

• Signal and noise characteristics are often nonstationary and the

statistical parameters vary with time
• Adaptive filter has an adaptation algorithm, that is meant to monitor the
environment and vary the filter transfer function accordingly
• Based on the actual inputs received, the algorithm attempts to find the
optimum filter design
 WORK PLAN

S.No TIME LINE WORK DONE/TO BE DONE

1 AUGUST-SEPTEMBER Analysis of Research Papers and Reference Book

2 SEPTEMBER-OCTOBER Learned about basic types of most occurring noises in general and removal of
noises in spatial and frequency domains

3 OCTOBER-NOVEMER Implementation of filtering techniques in Image Restoration(Coding Part)

4 NOVEMBER-DECEMBER Optimizing the basic algorithm such that it works adaptively for images
consisting of different kinds of noises
 CONCLUSION

• In this presentation definition and objectives of image restoration are addressed. In addition to
that, noise models and different filtering techniques are elaborated.
• It is concluded from the above explanations that usage of the techniques is governed by the
understanding, requirement and the standard of the output needed.
• Restoration Techniques are used to restore corrupted images to their original form
• Spatial domain techniques are particularly useful for removing random noise
• Frequency domain techniques are particularly useful for removing periodic noise
 REFERENCES

[1] B. Li and Z. Zhan, "Research on Motion Blurred Image Restoration," 2012 5th International Congress on Image
and Signal Processing, 2012, pp. 1307-1311, doi: 10.1109/CISP.2012.6469900.
[2] M. Yadav and N. Tiwari, "Performance Comparison of Image Restoration techniques using CNN and their
applications," 2021 5th International Conference on Computing Methodologies and Communication (ICCMC), 2021,
pp. 1146-1151, doi: 10.1109/ICCMC51019.2021.9418340.
[3] W. Zhang, L. Dong, X. Pan, P. Zou, L. Qin and W. Xu, "A Survey of Restoration and Enhancement for
Underwater Images," in IEEE Access, vol. 7, pp. 182259-182279, 2019, doi: 10.1109/ACCESS.2019.2959560.
[4] W. Luo, S. Duan and J. Zheng, "Underwater Image Restoration and Enhancement Based on a Fusion Algorithm
With Color Balance, Contrast Optimization, and Histogram Stretching," in IEEE Access, vol. 9, pp. 31792-31804,
2021, doi: 10.1109/ACCESS.2021.3060947.
[5] Shuo Sun, Application of fuzzy image restoration in criminal investigation, Journal of Visual Communication and
Image Representation, Volume 71, 2020, 102704, ISSN 1047-3203
[6] Sungjin Kim and Joonki Paik, "Fast video enhancement algorithm using real-time image restoration framework,"
Proceedings of 2002 IEEE 10th Digital Signal Processing Workshop, 2002 and the 2nd Signal Processing Education
Workshop., 2002, pp. 108-113, doi: 10.1109/DSPWS.2002.1231086.