Image pre processing-restoration

Image Pre-Processing:
Image Restoration
Ashish Khare

 Imagerestoration - suppressing image
degradation using knowledge about its nature

Image restoration as inverse
convolution of the whole image
 Most image restoration methods are based on
convolution applied globally to the whole image.

 Degradation causes:
 defects of optical lenses,
 nonlinearity of the electro-optical sensor,
 graininess of the film material,
 relative motion between an object and camera
 wrong focus,
 atmospheric turbulence in remote sensing or astronomy,
 etc.

 Theobjective of image restoration is to
reconstruct the original image from its
degraded version.

 Image restoration techniques - two groups:

 Deterministic methods - applicable to images with
little noise and a known degradation function.
 The original image is obtained from the degraded one by a
transformation inverse to the degradation.

 Stochastic techniques - the best restoration is
sought according to some stochastic criterion, e.g.,
a least squares method.
 In some cases the degradation transformation must be
estimated first.

 It is advantageous to know the degradation function
explicitly.

 The better this knowledge is, the better are the
results of the restoration.

 There are three typical degradations with a simple
function:
 Relative constant speed movement of the object with
respect to the camera,
 wrong lens focus,
 and atmospheric turbulence.

 Inmost practical cases, there is insufficient
knowledge about the degradation, and it must
be estimated and modeled.

 Methods:
 A priori knowledge about degradation - either known in
advance or obtained before restoration.
 If it is clear in advance that the image was degraded by
relative motion of an object with respect to the sensor then
the modeling only determines the speed and direction of
the motion.

 An example of parameters obtained before
restoration is an attempt to estimate parameters of a
capturing device such as a TV camera or digitizer.

 A posteriori knowledge is obtained by analyzing
the degraded image.

 A typical example is to find some interest points in
the image (e.g. corners, straight lines) and guess
how they looked before degradation.

 Another possibility is to use spectral characteristics
of the regions in the image that are relatively
homogeneous.

A degraded image g can arise from the
original image f by a process which can be
expressed as

where s is some nonlinear function and nu
describes the noise.

 The degradation is very often simplified by
 neglecting the nonlinearity
 assuming that the function h is invariant with
respect to position in the image.

Degradation can be then expressed as
convolution

 Ifthe degradation is given by above equation
and the noise is not significant then image
restoration equates to inverse convolution
(also called deconvolution).

 Ifnoise is not negligible then the inverse
convolution is solved as an overdetermined
system of linear equations.

Degradations that are
easy to restore
 Some degradations can be easily expressed
mathematically (convolution) and also
restored simply in images.

 TheFourier transform H of the convolution
function is used.

Relative motion of
the camera and object
 Assume an image is acquired with a camera
with a mechanical shutter.

 Relativemotion of the camera and the
photographed object during the shutter open
time T causes smoothing of the object in the
image.

 Suppose V is the constant speed in the
direction of the x axis; the Fourier transform
H(u,v) of the degradation caused in time T is
given by

Wrong lens focus
 Image smoothing caused by imperfect focus
of a thin lens can be described by the
following function

where J1 is the Bessel function of the first
order, r2 = u2 + v2, and a is the displacement.

Atmospheric turbulence
 needs to be restored in remote sensing and
astronomy
 caused by temperature non-homogeneity in the
atmosphere that deviates passing light rays.
 The mathematical model

where c is a constant that depends on the type of
turbulence which is usually found experimentally. The
power 5/6 is sometimes replaced by 1.

Inverse filtration
 based on the assumption that degradation
was caused by a linear function h(i,j)

 theadditive noise nu is another source of
degradation.

 Itis further assumed that nu is independent
of the signal.

 Applying the Fourier transform

 The degradation can be eliminated if the
restoration filter has a transfer function that is
inverse to the degradation h ... or in Fourier
transform H-1(u,v).

 Theundegraded image F is derived from its
degraded version G.

 Thisequation shows that inverse filtration
works well for images that are not corrupted
by noise.

 If noise is present and additive error occurs, its
influence is significant for frequencies where H(u,v)
has small magnitude. These usually correspond to
high frequencies u,v and thus fine details are blurred
in the image.
 The changing level of noise may cause problems as
well because small magnitudes of H(u,v) can cause
large changes in the result.
 Inverse filter values should not be of zero value so
as to avoid zero divides in above equation.

Wiener filtration
 Itis no surprise that inverse filtration gives
poor results in pixels suffering from noise
since the noise is not taken into account.

 Wiener filtration explores a priori knowledge
about the noise.

 Restoration by the Wiener filter gives an
estimate of the original uncorrupted image f
with minimal mean square error e2

 No constraints applied to above equation
 optimal estimate ^f is the conditional mean value
of the ideal image f under the condition g
 complicated from the computational point of view
 also, the conditional probability density between
the optimal image f and the corrupted image g is
not usually known.
 the optimal estimate is in general a nonlinear
function of the image g.

 Minimization of e2 is easy if estimate ^f is a
linear combination of the values in the image
g -close-optimal solution

H ... Fourier transform of the Wiener filter -
W
considers noise - see Eq. below
 G ... Fourier transform of the degraded image

 Where,
H is the transform function of the degradation
# denotes complex conjugate
S_{nu nu} is the spectral density of the noise
S_{gg} is the spectral density of the degraded
image

 If Wiener filtration is used, the nature of degradation
H and statistical parameters of the noise need to be
known.

 Wiener filtration theory solves the problem of a
posteriori linear mean square estimate -- all statistics
(e.g., power spectrum) should be available in
advance.

 Note that the inverse filter discussed earlier is a
special case of the Wiener filter where noise is
absent i.e. S_{nu nu} = 0.

Image pre processing-restoration

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Image pre processing-restoration (20)

More from Ashish Kumar (18)

Image pre processing-restoration