Title: Smoothing Techniques in Image Processing
1Smoothing Techniques in Image Processing
- Prof. PhD. Vasile Gui
- Polytechnic University of Timisoara
2Content
- Introduction
- Brief review of linear operators
- Linear image smoothing techniques
- Nonlinear image smoothing techniques
3Introduction
- Why do we need image smoothing?
- What is image and what is noise?
- Frequency spectrum
- Statistical properties
-
4Brief Review of Linear OperatorsPratt 1991
- Generalized 2D linear operator
- Separable linear operator
- Space invariant operator
-
- Convolution sum
5Brief Review of Linear Operators
- Geometrical interpretation of 2D convolution
6Brief Review of Linear Operators
- Matrix formulation
- Unitary transform
- Separable transform
- Transform is invertible
-
7Brief Review of Linear Operators
- Basis vectors are orthogonal
- Inner product and energy are conserved
- Unitary transform a rotation in N-dimensional
vector space
8Brief Review of Linear Operators
- 2D vector space interpretation
- of a unitary transform
9Brief Review of Linear Operators
10Brief Review of Linear Operators
11Brief Review of Linear Operators
- Circular convolution theorem
- Periodicity of f,h and g with N are assumed
12Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
13Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- Separable
- Can be computed recursively, resulting in roughly
4 operations per pixel
14Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- Optimality properties
- Signal and additive white noise
- Noise variance is reduced N times
15Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- Unknown constant signal plus noise
- Minimize MSE of the estimation g
16Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- i.i.d. Gaussian signal with unknown mean.
- Given the observed samples, maximize
- Optimal solution arithmetic mean
17Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- Frequency response
- Non-monotonically decreasing with frequency
18Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
19Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
- Image smoothed with 3?3, 5?5, 9?9
- and 11 ?11 box filters
20Linear Image smoothing techniques Box filters.
Arithmetic mean L?L operator
Lena image filtered with 5x5 box filter
Original Lena image
21Linear Image smoothing techniques Binomial
filters Jahne 1995
- Computes a weighted average of pixels in the
window - Less blurring, less noise cleaning for the same
size - The family of binomial filters can be defined
recursively - The coefficients can be found from (1x)n
22Linear Image smoothing techniques Binomial
filters. 1D versions
As size increases, the shape of the filter is
closer to a Gaussian one
23Linear Image smoothing techniques Binomial
filters. 2D versions
24Linear Image smoothing techniques Binomial
filters. Frequency response
Monotonically decreasing with frequency
25Linear Image smoothing techniques Binomial
filters. Example
Original Lena image
Lena image filtered with binomial 5x5 kernel
Lena image filtered with box filter 5x5
26Linear Image smoothing techniques Binomial and
box filters. Edge blurring comparison
- Linear filters have to compromise smoothing with
edge blurring
27Nonlinear image smoothingThe median filter
Pratt 1991
N is odd
28Nonlinear image smoothingThe median filter
29Nonlinear image smoothingThe median filter
median f1 f2 ? median f1 median
f2. However median c f c median f
, median c f c median f .
- The filter selects a sample from the window,
does not average - Edges are better preserved than with liner
filters - Best suited for salt and pepper noise
30Nonlinear image smoothingThe median filter
Noisy image
5x5 median filtered
5x5 box filter
31Nonlinear image smoothingThe median filter
- Optimality
- Grey level plateau plus noise. Minimize sum of
absolute differences - Result
- If 51 of samples are correct and 49 outliers,
the median still finds the right level!
32Nonlinear image smoothingThe median filter
- Caution points, thin lines and corners are
erased by the median filter
33Nonlinear image smoothingThe median filter
- Cross shaped window can correct some of the
problems
34Nonlinear image smoothingThe median filter
- Implementing the median filter
- Sorting needs O(N2) comparisons
- Bubble sort
- Quick sort
- Huang algorithm (based on histogram)
- VLSI median
35Nonlinear image smoothingThe median filter
- VLSI median block diagram
36Nonlinear image smoothingThe median filter
- Color median filter
- There is no natural ordering in 3D (RGB) color
space - Separate filtering on R,G and B components does
not guarantee that the median selects a true
sample from the input window - Vector median filter, defined as the sample
minimizing the sum of absolute deviations from
all the samples - Computing the vector median is very time
consuming, although several fast algorithms exist
37Nonlinear image smoothingThe median filter
- Example of color median filtering
- 5x5 pixels window
- Up original image
- Down filtered image
38Nonlinear image smoothingThe weighted median
filter
- The basic idea is to give higher weight to some
samples, according to their position with respect
to the center of the window - Each sample is given a weight according to its
spatial position in the window. - Weights are defined by a weighting mask
- Weighted samples are ordered as usually
- The weighted median is the sample in the ordered
array such that neither all smaller samples nor
all higher samples can cumulate more than 50 of
weights. - If weights are integers, they specify how many
times a sample is replicated in the ordered array
39Nonlinear image smoothingThe weighted median
filter
- Numerical example for the weighted median filter
40Nonlinear image smoothingThe multi-stage median
filter
- Better detail preservation
41Nonlinear image smoothingRank-order filters (L
filters)
42Nonlinear image smoothingRank-order filters (L
filters)
Percentile filters (rank selection) 0, 50 100
Particular cases of grey level morphological
filters
43Nonlinear image smoothingRank-order filters (L
filters)
Minkovski distance
Case r 1 best estimator is median. Case r 2,
best estimator is arithmetic mean. Case r ? ?,
best estimator is mid-range.
44Nonlinear image smoothingRank-order filters (L
filters)
- Alpha-trimmed mean filter
- ? (2Q 1) / N, defines de degree of averaging
- 1 corresponds to arithmetic mean
- ? 1 / N corresponds to the median filter
Properties in between mean and median
45Nonlinear image smoothingRank-order filters (L
filters)
- Median of absolute differences trimmed mean
- Better smoothing than the median filter and good
edge preservation
M2 is a robust estimator of the variance
46Nonlinear image smoothingRank-order filters (L
filters)
- k nearest neighbour (kNN) median filter
- Median of the k grey values nearest by rank to
the central pixel - Aim same as above
47Nonlinear image smoothingSelected area filtering
Nagao 1980
- Kuwahara type filtering
- Form regions Ri
- Compute mean, mi and variance, Si for each
region. - Result mi mi lt mj for all j different from i,
i,e. the mean of the most homogeneous region - Matsujama Nagao
48Nonlinear image smoothingConditional mean
- Pixels in a neighbourhood are averaged only if
they differ from the central pixel by less than a
given threshold
L is a space scale parameter and th is a range
scale parameter
49Nonlinear image smoothingConditional mean
50Nonlinear image smoothingBilateral filter
Tomasi 1998
- Space and range are treated in a similar way
- Space and range similarity is required for the
averaged pixels - Tomasi and Manduchi 1998 introduced soft
weights to penalize the space and range
dissimilarity.
s() and r() are space and range similarity
functions (Gaussian functions of the Euclidian
distance between their arguments).
51Nonlinear image smoothingBilateral filter
- The filter can be seen as weighted averaging in
the joint space-range space (3D for monochromatic
images and 5D x,y,R,G,B - for colour images) - The vector components are supposed to be properly
normalized (divide by variance for example) - The weights are given by
52Nonlinear image smoothingBilateral filter
- Example of Bilateral filtering
- Low contrast texture has been removed
- Yet edges are well preserved
53Nonlinear image smoothingMean shift filtering
Comaniciu 1999, 2002
- Mean shift filtering replaces each pixels value
with the most probable local value, found by a
nonparametric probability density estimation
method. - The multivariate kernel density estimate obtained
in the point x with the kernel K(x) and window
radius r is - For the Epanechnikov kernel, the estimated
normalized density gradient is proportional to
the mean shift
S is a sphere of radius r, centered on x and nx
is the number of samples inside the sphere
54Nonlinear image smoothingMean shift filtering
- The mean shift procedure is a gradient ascent
method to find local modes (maxima) of the
probability density and is guaranteed to
converge. - Step1 computation of the mean shift vector
Mr(x). - Step2 translation of the window Sr(x) by Mr(x).
- Iterations start for each pixel (5D point) and
tipically converge in 2-3 steps.
55Nonlinear image smoothingMean shift filtering
56Nonlinear image smoothingMean shift filtering
- Detail of a 24x40
- window from the
- cameraman image
- a) Original data
- b) Mean shift paths for some points
- c) Filtered data
- d) Segmented data
57Nonlinear image smoothingMean shift filtering
58Nonlinear image smoothingMean shift filtering
- Comparison to bilateral filtering
- Both methods based on simultaneous processing of
both the spatial and range domains - While the bilateral filtering uses a static
window, the mean shift window is dynamic, moving
in the direction of the maximum increase of the
density gradient.
59REFERENCES
- D. Comaniciu, P. Meer Mean shift analysis and
applications. 7th International Conference on
Computer Vision, Kerkyra, Greece, Sept. 1999,
1197-1203. - D. Comaniciu, P. Meer Mean shift a robust
approach toward feature space analysis. IEEE
Trans. on PAMI Vol. 24, No. 5, May 2002, 1-18. - M. Elad On the origin of bilateral filter and
ways to improve it. IEEE trans. on Image
Processing Vol. 11, No. 10, October 2002,
1141-1151. - B. Jahne Digital image processing. Springer
Verlag, Berlin 1995. - M. Nagao, T. Matsuiama A structural analysis of
complex aerial photographs. Plenum Press, New
York, 1980. - W.K. Pratt Digital image processing. John Wiley
and sons, New York 1991 - C. Tomasi, R. Manduchi Bilateral filtering for
gray and color images. Proc. Sixth Intl. Conf.
Computer Vision , Bombay, 839-846.