Vous êtes sur la page 1sur 4

Project Title:

Motion Deblurring

Project Supervisor:
Dr. Vandana Dixit Kaushik
Mrs. Monika Verma

Group Members (FINAL CSE):

APARNA BHATT (91/11)


DEEPAK KUMAR (226/11)
RAGHAV MISHRA(382/11)
SHIVAM GUPTA(262/11)

Aim :
To design enhanced techniques and algorithms for deblurring the

motion blurred images.

Objective:
To analyze how the various images parameters get destroyed/ blurred due to
motion and
developing algorithms using existing techniques, algorithms,
research work and approaches from computer vision, artificial intelligence but
particularly digital image processing to find its role and contribution in blurred
images analysis.

Description and Importance


Motion blur is the result of the relative motion between the camera and the
original scene during the integration time of the image. It can be clearly seen in
the pictures that were taken with long exposure times and the pictures of fast
moving objects. The motion blur effect is often used in computer graphics to
make synthetic images and animations look more realistic and to add additional
information about the art and the direction of the motion. On the other hand, the
real-world images often suffer from very strong motion blurs. The typical reasons
are the camera motion and the motion of an object on the scene.
Motion blur is not always a negative effect. It makes images and animations look
more realistic in certain situations. Motion blur can be observed in almost every
movie and TV program. This effect is simulated in many computer games.
Motion blurred images contain additional information about the direction of the
movement and are more pleasing to our eyes.
Restoration of blurred images is very important problem in tracking and
identification of criminals ,where image of a human face or number plate of a
running vehicle taken in hit and run situation gets blurred due to relative motion
between the imaging system and object/face.

Proposed Methodology:
The process of motion deblurring can be divided into two parts: the estimation of
the function that
caused the blur, and applying a restoration algorithm. Since the motion path can
be arbitrary, the first problem can be hard to solve.
In image deblurring , we seek to recover the original, sharp image by using a
mathematical model of the blurring process. The key issue is that some
information on the lost details is indeed present in the blurred imagebut this
information is hidden and can only be recovered if we know the details of the
blurring process.
Unfortunately there is no hope that we can recover the original image exactly!
This is due to various unavoidable errors in the recorded image. The most
important errors are fluctuations in the recording process and approximation
errors when representing the image with a limited number of digits. The
influence of this noise puts a limit on the size of the details that we can hope to
recover in the reconstructed image, and the limit depends on both the noise and
the blurring process.

Algorithms and techniques to be employed:


Mathematical Models
In order to solve the image restoration problem, mathematical models are
required for the real world processes involved in image generation, formation
and recording. The following subsections we will learn the basic concepts of
convolution, point spread function and spatial invariance. Afterwards, we will
derive the equation for the image formation process.

2-D Convolution
Image Formation
Point Spread Function

For restoration of motion blurred images, knowledge of the point spread function
(PSF) is very important. The motion blur PSF is characterized by two parameters,
namely blur direction and blur length which are used to restore images from the
motion blurred images.

Deconvolution
The process of reconstruction of the original image from the blurred one with the
same blurring function is called deconvolution. Knowing the PSF and the nature
of noise, it may be possible to (at least partially) restore the ideal image. The
degree to which this can be done depends on the actual sampling of the image.
Various algorithms and techniques are to be used under Deconvolution. Some of
them are:
Inverse Filtering

Weiner Filter based algorithm


Iterative Image restoration
Richardson-Lucy algorithm

Hybrid Imaging Approach to Motion Deblurring


It includes
Hybrid Imaging Systems
Computing Motion
PSF Estimation
Deblurring of Moving Objects
Few definitions:
Weiner filter based algorithm:
A Weiner filter based algorithm is to restore the images from motion blurred and
noisy images has also been presented.
The experimental results have shown that (i) the parameter of Weiner filter can
be approximated to the reciprocal of image width, (ii) in artificially blurred
images the identified blur length and blur direction is very close to theoretical
value, and (iii) the proposed image restoration technique effectively restores the
images from natural motion blurred, artificial motion blurred and noisy images.
Patch based algorithm:
Using this mechanism, we will develop a fast and robust deblurring technique
based on patch learning mechanism from the input noisy blur image itself.
Robustness and efficiency issues require a much detailed and careful analysis.
It is shown experimentally that while the proposed method can produce
comparable or better result compared to state-of-the-art methods and it is 3
times faster than existing ones. However, the proposed method is not suitable
for deblurring under asymmetric/ sparse kernel. In this work we would explore
those issues in detail and hopefully address other inverse methods for imaging
as well.

Tools and Platform to be used:


MATLAB is an excellent environment in which to develop and experiment with
filtering Methods for image deblurring. The basic MATLAB package contains
many functions and tools for this purpose, but in some cases it is more
convenient to use routines that are only available from the Signal Processing
Toolbox (SPT) and the Image Processing Toolbox (IPT).We will therefore use these
toolboxes when convenient.
The Image Processing Toolbox is a collection of MATLAB functions (called Mfunctions or M-files) that extend the capability of the MATLAB environment for
the solution of BLURRED IMAGE problems. Other toolboxes that sometimes are
used to complement the Image Processing Toolbox are the Signal Processing,
Neural Networks, Fuzzy Logic, and Wavelet Toolboxes.

SYNOPSIS
(MOTION DEBLURRING)

Vous aimerez peut-être aussi