1

CHAPTER 1
INTRODUCTION
Medical image enhancement technologies have attracted much attention
since advanced medical equipments were put into use in the medical field.
Enhanced medical images are desired by a surgeon to assist diagnosis and
interpretation because medical image qualities are often deteriorated by noise and
other data acquisition devices, illumination conditions, etc. Also targets of medical
image enhancement are mainly to solve problems of low contrast and the high
level noise of a medical image.
Medical image enhancement technologies have attracted many studies,
mainly on greyscale transform and frequency domain transform. Studies of
frequency domain transform mainly concentrate on the wavelet transform, and
histogram equalisation is a quite typical method of image enhancement in the
spatial field. The wavelet transform is a time-frequency analysis tool developed in
the 1980s, which has been successfully applied in the image processing domain
after Mallat presented the fast decomposition algorithm. There are many image
enhancement methods based on wavelet transform. In these papers, methods of
image enhancement based on wavelet transform were proposed. However, we
cannot obtain more high-frequency information only through multi-scale wavelet
transform. An images different scale detail information can be obtained through
wavelet transform, but there will be some high-frequency information hidden in
high-frequency sub-images of wavelet transform. If we decompose these high-
frequency sub-images, we can obtained more image high-frequency information
which can help us to enhance a medical image effectively. Also, we can obtain a
better enhancement image if we use both spatial field and transform field
procession to enhance an image. In addition, we should remove or reduce noise
for the reason that there is lots of noise in high-frequency sub-images. Presented
in this Letter is a novel approach which is used to enhance a medical image based
on wavelet transform, Haar transform and nonlinear histogram equalisation.
1.1 OBJECTIVE
In this project used to enhance the medical image. I use two level wavelet
transform here. In the step of the nonlinear histogram equalisation, we set M
2

255/3 and N f max/4. The enhancement weight coefficient is available from the
experimental results. The original images are converted by the proposed algorithm
then I can get the histogram equalisation results. The PSNR value of proposed
algorithm is 39.64, and of histogram equalisation is 30.26, the PSNR value of
proposed algorithm another value is 70.53, and that of another value of histogram
equalisation is 45.53. From the enhanced results, enhanced images with the
method proposed are better than results with histogram equalisation. In proposed
algorithm results, not only are the images blur and low-contrast detail enhanced,
but the images textures are clearer.
1.2 EXISTING SYSTEM
Medical image enhancement technologies have attracted many studies,
mainly on gray scale transform and frequency domain transform. The frequency
domain transform mainly concentrate on the wavelet transform, and histogram
equalization is a quite typical method of image enhancement in the spatial field.

1.2.1 Disadvantages of Existing System
Image processing and analysis based on the continuous or discrete image
transforms are classic techniques. The image transforms are widely used in image
filtering, data description, etc. The wavelet theorems make up very popular
methods of image processing, denoising and compression. The image transform
theory is a well known area characterized by a precise mathematical background,
but in many cases some transforms have particular properties which are not still
investigated. So I present a method of image analysis by means of the wavelets
Haar spectrum.

1.2.2 Literature Survey
In the title of project Multifrequency channel decompositions of image and
wavelet models, by Mallat, S.G.: IEEE Trans. Acoustic. Speech Signal Process.,
1989, 37, (12), pp. 20912110, described the application of multifrequency
decompositions to images processing from several viewpoints.It covered some
psycgophysical and physiological data showing that such a decomposition seems .
It first reviewed the properties of a window fourier transform and explained why
this decomposition is not convenient for analyzing signals such as images. Then
3

introduced the wavelet transform and described its most importatnt properties.
Although the goal of this paper was not build any psychophysicalogical model to
some low level processes in the visual cortex.
A uniform sampling of each multi frequency channel defines a representation
which is not translation invariant. It is therefore difficult to build pattern recognition
algorithms from such decompositions. We reviewed the properties of zero crossing
in multiband decompositions. This adaptive sampling is translation invariant but is
much more difficult to analyze.
Gray and Colour Image Contrast Enhancement by the Curve let Transform
by Jean-Luc Starck, Fionn Murtagh, Emmanuel J. Cands, and David L. Donoho
from IEEE transactions on image processing, vol. 12, no. 6, june 2003 from the
paper A number of properties, respected by the curve let filtering described here,
are important for contrast stretching. Reconstruct the enhanced image from the
modified curve let coefficients. Noise must not be amplified in enhancing edges.
Reconstruct the enhanced image from the modified curve let coefficients. Colours
should not be unduly modified. In the multiscale retinex, for example, a tendency
toward increased grayness is seen. This is not the case using curvelets. However
colour restoration could also be carried out in a final step, as proposed for the
multistage retinex [1]. This should improve the final image quality. Reconstruct the
enhanced image from the modified curvelet coefficients. It is very advantageous if
block effects do not occur. Block overlapping is usually not necessary in curvelet-
based contrast enhancement, unlike in the case of noise filtering. A range of
further examples can be seen at Our conclusions are as follows. The curvelet and
wavelet enhancement functions take account very well of image noise. As
evidenced by the experiments with the curvelet transform, there is better detection
of noisy contours than with other methods. For noise-free images, there is not a
great deal to be gained by curvelet enhancement over wavelet enhancement since
the enhancement function tends toward Veldes approach in such weak noise
cases. Contours and edges are detected quite adequately by wavelets in such
situations.
Contrast enhancement of colour images Based on wavelet transform and
human visual system by Ding Xiao Jun Ohya , proceedings of the IASTED
International Conference This paper has proposed a colour contrast enhancement
method that uses a luminance component enhancement based on wavelet
4

transform: more specifically, Reverse- S-Shape enhancement based on human
visual system for the approximate component coefficients obtained by the Wavelet
transform. The Saturation components are enhanced by histogram equalization. It
turns out that the proposed wavelet based colour contrast enhancement method
can achieve a successful enhancement of colour images which are dark or with
low contrast. However, there are still some remaining issues. The transformation
algorithm for the approximate coefficients is to be improved and the enhancement
of detail coefficients may also be effective. The relationship between luminance
value and saturation is not considered in the method. The method is a global
transformation of a certain image but the performance might be better if we divide
the image into some certain areas according to some certain rules and apply a
different algorithm or different parameters to different areas. Another topic is that
sometimes the colour contrast enhancement requires changing colour; that is, the
hue component should also be adjusted.
Contrast enhancement of medical images using multi-scale edge
representation, Opt. Eng., 1994, 33, (7), pp. 21512161by Lu, J., Healy, D.M.,
and Weaver, J.B , in this paper presented a description of an edge- base contrast
enhancement method , and demonstrated its use with images drawn from three
important medical imagining modalities.
Edge-based enhancement offers certain attractive advantages over histogram
based techniques. For examples we want contrast enhancement to take place
along the direction perpendicular to edges in the image, and histogram based
techniques do not provide such spatial and orientation localization.
The concept of edge-based contrast enhancement is not completely new:
however, previously developed techniques all techniques all experiences certain
difficulties in implementing the basic idea. We have presented some evidence that
many of these difficulties can be addressed in an effective way by using wavelet
based multiscale edge representation of images. Additionally , the wavelet
approach offers , among other advantages to be further explored , flexibility to
selectively enhance features of different sizes and in different locations , as well
as the ability to control noise magnification.
New algorithm of image enhancement based on wavelet transform, Comput.
Appl., 2005, 25, (3) by Zhou, X., Zhou, S., Huang, F., and Zhou, X.T image
enhancement is a technology to improve the quality of an image in terms of visual
5

perception of human beings. With the growing quality in image acquisition, image
enhancement technologies are more and more needed for many applications.
Images are categorized into grey-level images and colour images. Each pixel of
the grey-level image has only one grey-level value as opposed to colour images
pixels; therefore, there have been many algorithms for contrast enhancement for
grey-level images. The main techniques for image enhancement such as contrast
stretching, slicing, histogram equalization, for grey-level images are discussed in
many articles and books. On the other hand, since each pixel of colour images
consists of colour information as well as grey-level information, these typical
techniques for grey-level images cannot be applied to colour images. Thus,
compared with grey-level images, the enhancement of colour images is more
difficult, and there are much more points to be researched.

1.3 PROPOSED SYSTEM
In this medical image is decomposed with wavelet transform. Then, all high
frequency sub-images were decomposed with Haar transform. Noises in the
frequency fields are reduced by the soft-threshold method High-frequency
coefficients were enhanced by different weight values in different sub-images.
Then, the enhanced image was obtained through the inverse wavelet transform
and inverse Haar transform.

1.3.1 Advantages
Medical image enhancement based on wavelet transform is how to extract
high-frequency information. Haar transform is used to decompose the high-
frequency sub-images of wavelets in this topic. This helps us to extract high-
frequency information effectively. Different enhancement weight coefficients in
different sub-images and nonlinear histogram equalisation are used in the process
of medical image enhancement. They can also help us to enhance a medical
image effectively. Results of experiments show that the algorithm not only can
enhance an images contrast, but also can preserve the original images edge
property effectively.



6

CHAPTER 2
ENHANCE AN IMAGE BY PROPOSED SYSTEM
Image enhancement process consists of a collection of technique that seek to
improve visual appearance of an image basically we have two different methods
they are
spatial domain
frequency domain
2.1 NEED OF THE TRANSFORM
Mathematical transformations are applied to signals to obtain further
information from that signal that is not readily available in the raw signal. In the
following tutorial I will assume a time-domain signal as a raw signal, and a signal
that has been "transformed" by any of the available mathematical transformations
as a processed signal
Most of the signals in practice are TIME-DOMAIN signals in their raw format.
That is, whatever that signal is measuring, is a function of time. In other words,
when we plot the signal one of the axes is time (independent variable), and the
other (dependent variable) is usually the amplitude. When we plot time-domain
signals, we obtain a time-amplitude representation of the signal. This
representation is not always the best representation of the signal for most signal
processing related applications. In many cases, the most distinguished information
is hidden in the frequency content of the signal. The frequency spectrum of a
signal is basically the frequency components (spectral components) of that signal.
The frequency spectrum of a signal shows what frequencies exist in the signal.
Intuitively, we all know that the frequency is something to do with the change in
rate of something. If something (a mathematical or physical variable, would be the
technically correct term) changes rapidly, we say that it is of high frequency, where
as if this variable does not change rapidly, i.e., it changes smoothly, we say that it
is of low frequency. If this variable does not change at all, then we say it has zero
frequency, or no frequency. For example the publication frequency of a daily
newspaper is higher than that of a monthly magazine (it is published more
frequently).
The frequency is measured in cycles/second, or with a more common name,
in "Hertz". For example the electric power we use in our daily life in the US is 60
7

Hz (50 Hz elsewhere in the world). This means that if you try to plot the electric
current, it will be a sine wave passing through the same point 50 times in 1
second. Now, look at the following figures. The first one is a sine wave at 3 Hz, the
second one at 10 Hz, and the third one at 50 Hz. Compare them.


Fig: 1 various frequency spectrums
So how do we measure frequency, or how do we find the frequency content of
a signal? The answer is FOURIER TRANSFORM (FT).
The frequency axis starts from zero, and goes up to infinity. For every
frequency, we have an amplitude value. For example, if we take the FT of the
electric current that we use in our houses, we will have one spike at 50 Hz, and
nothing elsewhere, since that signal has only 50 Hz frequency component. No
other signal, however, has a FT which is this simple. For most practical purposes,
signals contain more than one frequency component. The following shows the FT
of the 50 Hz signal


8

2.2 FREQUENCY DOMAIN
Let's give an example from biological signals. Suppose we are looking at an
ECG signal (Electro Cardiograph, graphical recording of heart's electrical activity).
The typical shape of a healthy ECG signal is well known to cardiologists. Any
significant deviation from that shape is usually considered to be a symptom of a
pathological condition.

2.3 BLOCK DIAGRAM OF PROPOSED SYSTEM
Frequency domain processing technique is based on modifying the
transformed of an image. To enhance the image in frequency domain method we
should follow the following steps
1. Wavelet transforms
2. Haar transform
3. Soft threshold method
4. Different weight values
5. Histogram equalization





Fig: 2 Block diagram of proposed system

2.4 WAVELT TRANSFORM
The wavelet transform is a transform of this type. It provides the time-
frequency representation. (There are other transforms which give this information
too, such as short time Fourier transform, Wigner distributions, etc.)
Often times a particular spectral component occurring at any instant can be of
particular interest. In these cases it may be very beneficial to know the time
intervals these particular spectral components occur. For example, in EEGs, the
latency of an event-related potential is of particular interest (Event-related potential
is the response of the brain to a specific stimulus like flash-light, the latency of this
response is the amount of time elapsed between the onset of the stimulus and the
response).
9


Fig: 3 Comparison of frequency and wavelet transform
Wavelet transform is capable of providing the time and frequency information
simultaneously, hence giving a time-frequency representation of the signal.
How wavelet transform works is completely a different fun story, and should be
explained after short time Fourier Transform (STFT). The WT was developed as
an alternative to the STFT. The STFT will be explained in great detail in the
second part of this tutorial. It suffices at this time to say that the WT was
developed to overcome some resolution related problems of the STFT, as
explained in Part II.
To make a real long story short, we pass the time-domain signal from various
high pass and low pass filters, which filter out either high frequency or low
frequency portions of the signal. This procedure is repeated, every time some
portion of the signal corresponding to some frequencies being removed from the
signal.
Here is how this works: Suppose we have a signal which has frequencies up
to 1000 Hz. In the first stage we split up the signal in to two parts by passing the
signal from a highpass and a lowpass filter (filters should satisfy some certain
10

conditions, so-called admissibility condition) which results in two different
versions of the same signal: portion of the signal corresponding to 0-500 Hz (low
pass portion), and 500-1000 Hz (high pass portion).
Then, we take either portion (usually low pass portion) or both, and do the same
thing again. This operation is called decomposition. Assuming that we have taken
the low pass portion, we now have 3 sets of data, each corresponding to the same
signal at frequencies 0-250 Hz, 250-500 Hz, 500-1000 Hz.
Then we take the low pass portion again and pass it through low and
high pass filters; we now have 4 sets of signals corresponding to 0-125 Hz, 125-
250 Hz,250-500 Hz, and 500-1000 Hz. We continue like this until we have
decomposed the signal to a pre-defined certain level. Then we have a bunch of
signals, which actually represent the same signal, but all corresponding to different
frequency bands. We know which signal corresponds to which frequency band,
and if we put all of them together and plot them on a 3-D graph, we will have time
in one axis, frequency in the second and amplitude in the third axis. This will show
us which frequencies exist at which time (there is an issue, called "uncertainty
principle", which states that, we cannot exactly know what frequency exists at what
time instance, but we can only know what frequency bands exist at what time
intervals, more about this in the subsequent parts. Will be made more clear in the
subsequent sections, but it should be noted at this time that the scale is inverse of
frequency.
That is, high scales correspond to low frequencies, and low scales correspond
to high frequencies. Consequently, the little peak in the plot corresponds to the
high frequency components in the signal, and the large peak corresponds to low
frequency components (which appear before the high frequency components in
time) in the signal.
11


Fig:4 wavelet transform in three dimensions graph

2.4.1 Continuous Wavelet Transform

Fig: 5 wavelet transform in Translation

2.4.2 Multi resolution analysis
Although the time and frequency resolution problems are results of a physical
phenomenon (the Heisenberg uncertainty principle) and exist regardless of the
transform used, it is possible to analyze any signal by using an alternative
approach called the multi resolution analysis (MRA). MRA, as implied by its
12

name, analyzes the signal at different frequencies with different resolutions. Every
spectral component is not resolved equally as was the case in the STFT.
MRA is designed to give good time resolution and poor frequency resolution at
high frequencies and good frequency resolution and poor time resolution at low
frequencies. This approach makes sense especially when the signal at hand has
high frequency components for short durations and low frequency components for
long durations. Fortunately, the signals that are encountered in practical
applications are often of this type. For example, the following shows a signal of
this type. It has a relatively low frequency component throughout the entire signal
and relatively high frequency components for a short duration somewhere around
the middle.



Fig: 6 Analysis of multi resolution graph


2.4.3 The scale
Duration of the signal, unlike those shown in the figure, but they usually
appear from time to time as short bursts, or spikes. High scales (low frequencies)
usually last for the entire duration of the signal. Scaling, as a mathematical
13

operation, either dilates or compresses a signal. Larger scales correspond to
dilated (or stretched out) signals and small scales correspond to compressed
signals.

Fig: 7 Different Frequency scale
All of the signals given in the figure are derived from the same cosine signal, i.e.,
they are dilated or compressed versions of the same function. In the above figure,
s=0.05 is the smallest scale, and s=1 is the largest scale.

2.4.4 Computation of the CWT
Interpretation of the above equation will be explained in this section. Let x(t) is
the signal to be analyzed. The mother wavelet is chosen to serve as a prototype
for all windows in the process. All the windows that are used are the dilated (or
compressed) and shifted versions of the mother wavelet. There are a number of
functions that are used for this purpose. The Morlet wavelet and the Mexican hat
function are two candidates, and they are used for the wavelet analysis
Once the mother wavelet is chosen the computation starts with s=1 and the
continuous wavelet transform is computed for all values of s, smaller and larger
than ``1''. However, depending on the signal, a complete transform is usually not
necessary. For all practical purposes, the signals are band limited, and therefore,
computation of the transform for a limited interval of scales is usually adequate. In
this study, some finite interval of values for s was used.

14


If the signal has a spectral component that corresponds to the current value of
s (which is 1 in this case), the product of the wavelet with the signal at the
location where this spectral component exists gives a relatively large value. If
the spectral component that corresponds to the current value of s is not present in
the signal, the product value will be relatively small, or zero. The signal in Figure
3.3 has spectral components comparable to the window's width at s=1 around
t=100 ms.
The continuous wavelet transform of the signal in Figure will yield large values for
low scales around time 100 ms, and small values elsewhere. For high scales, on
the other hand, the continuous wavelet transform will give large values for almost
the entire duration of the signal, since low frequencies exist at all times.

Fig: 8 Intersection of wavelet and signal

2.4.5 Mathematical approach
This section describes the main idea of wavelet analysis theory, which can
also be considered to be the underlying concept of most of the signal analysis
techniques. The FT defined by Fourier use basis functions to analyze and
15

reconstruct a function. Every vector in a vector space can be written as a
linear combination of the basis vectors in that vector space, i.e., by
multiplying the vectors by some constant numbers, and then by taking the
summation of the products. The analysis of the signal involves the estimation of
these constant numbers (transform coefficients, or Fourier coefficients, wavelet
coefficients, etc). The synthesis, or the reconstruction, corresponds to computing
the linear combination equation.
This pathological condition, however, may not always be quite obvious in the
original time-domain signal. Cardiologists usually use the time-domain ECG
signals which are recorded on strip-charts to analyze ECG signals. Recently, the
new computerized ECG recorders/analyzers also utilize the frequency information
to decide whether a pathological condition exists. A pathological condition can
sometimes be diagnosed more easily when the frequency content of the signal is
analyzed

2.5 WAVELET TOOL BOX
Wavelet Toolbox extends the MATLAB technical computing environment with
graphical tools and command-line functions for developing wavelet-based
algorithms for the analysis, synthesis, denoising, and compression of signals and
images. Wavelet analysis provides more precise information about signal data
than other signal analysis techniques, such as Fourier. Wavelet Toolbox supports
the interactive exploration of wavelet properties and applications. It is useful for
speech and audio processing, image and video processing, biomedical imaging,
and one-dimensional (1-D) and two-dimensional (2-D) applications in
communications and geophysics.
2.5.1 Key features
Standard wavelet families, including Daubechies wavelet filters, complex
Morlet and Gaussian, real reverse
biorthogonal, and discrete Meyer
Wavelet and signal processing utilities, including a function to convert scale to
frequency
Methods for adding wavelet families
Lifting methods for constructing wavelets
Customizable presentation and visualization of data
16

Interactive tools for continuous and discrete wavelet analysis
Wavelet packets, implemented as MATLAB objects
One-dimensional multisignal analysis, compression, and denoising
Multiscale principal component analysis
Multivariate denoising

2.5.2 Applying wavelet methods
Wavelet methods provide powerful tools for analyzing, encoding,
compressing, reconstructing, and modeling
signals and images. They are useful in capturing, identifying, and analyzing
local, multiscale, and nonstationary
processes, enabling you to explore aspects of data that other analysis
techniques miss, such as trends, breakdown
points, discontinuities in higher derivatives, and self-similarity.

Fig: 9 Way of Applying Images
2.5.3 Wavelet Toolbox supports:
Enhance edge detection in image processing
Achieve high rates of signal or image compression with virtually no loss of
significant data
17

Restore noisy signals and degraded images
Discover trends in noisy or faulty data
Study the fractal properties of signals and images
Extract information-rich features for use in classification and pattern
recognition applications
Perform multivariate denoising of signals with multiscale principal component
analysis



Fig: 10 Analysing the signal to Haar Transform

2.5.4 Analyzing Signals and Images
The Wavelet Toolbox graphical user interface (GUI) provides a
comprehensive set of tools for analyzing 1-D and 2-D signals, including tools for
wavelet analysis, wavelet packet analysis, denoising, and compression. For 1-D
signals, you can use the GUI tools to:
18

Perform discrete wavelet analysis of signals
Perform continuous wavelet analysis of real signals using complex wavelets
Denoise signals
Estimate wavelet-based density
Perform wavelet reconstruction schemes based on various wavelet coefficient
selection strategies
Randomly generate fractional Brownian motion
Perform 1-D signal extension and truncation using periodic, symmetric,
smooth, and zeropadding methods
Perform 1-D signal clustering and classification using wavelet analyses (with
Statistics Toolbox, available
separately)
For 2-D signals, you can use the GUI tools to:
Perform discrete wavelet analysis of images
Fuse two images
Perform translation-invariant denoising of images, using the stationary
wavelet transform

2.5.4.1 Wavelet Based Image Coding

ENEE631 emphasis

Focus on conceptual aspects related to image compression
Wavelet is also useful for denoising, enhancement, and image analysis
Build upon filterbank and subband coding from ENEE630
(For more in-depth info. on wavelet: wavelet course offered in Math Dept.)

K-level 1-D wavelet/subband decomposition

Successive lowpass/highpass filtering and downsampling
on different level: capture transitions of different frequency bands
on the same level: capture transitions at different locations


19

2.5.5 Successive Wavelet/Sub band Decomposition


Fig: 11Successive decomposition in wavelet
Successive low pass/high pass filtering and down sampling on different level:
capture transitions of different frequency bands on the same level: capture
transitions at different locations


Fig: 12 Successive decomposition of images in wavelet

Then we can further subdivided by haar transform and the we can improve
the contrast of the picture and will get good resolution

20


Fig: 13 2-D Wavelet Transform via Separable Filters
The example of the 2-D as shown in th e following figure and as well as its
shows the decomposition method in the another diagram .

Fig: 14 Decomposition of original Image





Fig: 15 Decomposition by Level
21

2.5.5.1 Sub band Coding Techniques
General coding approach
Allocate different bits for coefficient. in different frequency bands
Encode different bands separately
Example: DCT-based JPEG and early wavelet coding
Some difference between subband coding and early wavelet coding ~ Choices of
filters
Subband filters aims at (approx.) non-overlapping freq. response
Wavelet filters has interpretations in terms of basis and typically designed for
certain smoothness constraints

Fig: 16 Sub band decomposition by Haar Transform
Shortcomings of sub band coding
Difficult to determine optimal bit allocation for low bit rate applications
Not easy to accommodate different bit rates with a single code stream
Difficult to encode at an exact target rate
2.5.6 Smoothness Conditions on Wavelet Filter
Ensure the low band coefficients obtained by recursive filtering can
provide a smooth approximation of the original signal Modern lossy wavelet
coding exploits multi-resolution and self-similar nature of wavelet decomposition.
22

Energy is compacted into a small number of coefficient. Significant coefficient.
Tend to cluster at the same spatial location in each frequency subband

2.5.6.1 EZW Algorithm and Example
Initial threshold ~ 2 ^ floor(log2 xmax)
Put all coeff. in dominant list
Dominant Pass (zig-zag across bands)
Assign symbol to each coeff. and
entropy encode symbols
ps positive significance
ns negative significance
iz isolated zero
ztr zero-tree root
Significant coeff.
Subordinate Pass
Output one bit for subordinate list
According to position in up/low half of quantization interval
Table 1: Decomposition matrix
















53 -22 21 -9 -1 8 -7 6
14 -12 13 -11 -1 0 2 -3
15 -8 9 7 2 -3 1 -2
34 -2 -6 10 6 -4 4 -5
-6 5 -1 1 1 3 -1 5
6 1 3 0 -2 2 6 0
4 2 1 -4 -1 0 -1 4
0 -2 7 5 -3 2 -2 3
23

After 1st Pass the output will be calculated by the below method
Table 2: Output of the first Dominant
Resulting output of the first dominant pass( to=32)
Sub band Co-efficient
value
Symbol Reconstruction
value
LL3 53 ps 48
HL3 -22 ztr 0
LH3 14 iz 0
HH3 -12 ztr 0
LH2 15 ztr 0
LH2 -8 ztr 0
LH2 34 ps 48
LH2 -2 ztr 0
LH1 4 iz 0
LH1 2 iz 0
LH1 0 iz 0
LH1 -2 iz 0


2.5.6.2 beyond EZW
Cons of EZW, Poor error resilience and Difficult for selective spatial
decoding. SPIHT (Set Partitioning in Hierarchal Trees) in this Further improvement
over EZW to remove redundancy. EBCOT (Embedded Block Coding with Optimal
Truncation) Used in JPEG 2000.

2.5.7 JPEG 2000 Image Compression Standard

Targets and features Excellent low bit rate performance without sacrifice
performance at higher bit rate. Progressive decoding to allow from lossy to
lossless. Region-of-interest (ROI) coding. Error resilience
24









Fig: 17Block Diagram of jpeg 2000 Decoder and Encoder

2.6 HAAR TRANSFORM
Fundamental of Haar Transform
In mathematics, the Haar wavelet is a certain sequence of rescaled
"square-shaped" functions which together form a wavelet family or basis. Wavelet
analysis is similar to Fourier analysis in that it allows a target function over an
interval to be represented in terms of an orthonormal function basis. The Haar
sequence is now recognised as the first known wavelet basis and extensively used
as a teaching example in the theory of wavelets. The haar transform is based on
the haar function, h
k
.These functions are defined over the continuous, Closed
interval z [0,1] and for k=0,1,2,.N-1 , where N=2
n
2.6.1 Construction of Haar functions
The Unique decomposition of integer k (p, q)
The value of k = 0, , N-1 with N = 2n, 0 <= p <= n-1
And the q = 0, 1 (for p=0); 1 <= q <= 2p (for p>0)

e.g., k=0 k=1 k=2 k=3 k=4
(0,0) (0,1) (1,1) (1,2) (2,1)


k = 2
p
+ q
remind
power
25

1
x
The basic function of haar Transform is given by the following equation which as
follows







for the corresponding values of the haar transforms function that the graph will be
developed













Fig: 18 Waveform of Haar Transform
2.6.2 Properties of Haar Transforms
The Haar wavelet has several notable properties:
1. Any continuous real function can be approximated by linear combinations of
and their shifted
functions. This extends to those function spaces where any function therein can be
approximated by continuous functions.

<
<
= =
= =
] 1 , 0 [ other for 0
2 2
for 2
1
2 2
1
for 2
1
) ( ) (
] 1 , 0 [ for
1
) ( ) (
2
1
2 /
2
1
2 /
,
0 , 0 0
x
q
x
q-
N
q-
x
q-
N
x h x h
x
N
x h x h
p p
p
p p
p
q p k
26

2. Any continuous real function can be approximated by linear combinations of
the constant function, and
their shifted functions.
3. Orthogonality in the form

Here
i,j
represents the Kronecker delta. The dual function of (t) is (t) itself.
4. Wavelet/scaling functions with different scale m have a functional relationship:
(t) = (2t) + (2t 1)
(t) = (2t) (2t 1)
5. Coefficients of scale m can be calculated by coefficients of scale m+1:
If
and
then



Its a real and orthogonal. Transition at each scale p is localized according to q
The value of the 8x8 matrix values derived using the basic equaition and the value
showed by







The Basis images of 2-D (separable) Haar transform which is
show by the following figure. And the Outer product of two basis vectors

(
(
(
(
(
(
(
(
(
(

2 2 0 0
0 0 2 2
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
2 2 0 0
0 0 2 2
2 2 2 2
0 0 0 0
1 1 1 1
1 1 1 1
0 0 0 0
2 2 2 2
1 1 1 1
1 1 1 1
8
1
27








Fig: 19 8x8 Haar basis Images

Now we can go for the comparison basic of images of DCT and Haar. The
Comparison diagram as follows.
In the haar Transform having the two major sub operators and that will be
described as follows .
1. Scaling captures info. at different frequencies
2. Translation captures info. at different locations

It Can be represented by filtering and downsampling and the Relatively poor
energy compaction. The wave representation of the final value showed in the
below figure.

Fig: 20 Waveform of Down sampling and 8x8 Haar basis images



1
x
28

Orthonormal Filter
Equivalent to projecting input signal to orthonormal basis. Energy
preservation property. It Convenient for quantizes design MSE by transform
domain quantized is same as reconstruction MSE.
Shortcomings: coefficient expansion. Its Linear filtering with N-element
input & M-element filter Which is described by the (N+M-1)-element output,
(N+M)/2 after down sample. The Length of output per stage grows is equal
undesirable for compression. The Solutions to coefficient expansion is
symmetrically extended input (circular convolution) & Symmetric filter









Fig: 21 symmetric and periodic extension of the image
Solutions to Coefficient Expansion
Circular convolution in place of linear convolution. The Periodic extension of
input signal and Problem: artifacts by large discontinuity at borders. Symmetric
extension of input is Reduce border artifacts (note the signal length doubled with
symmetry) and the Problem is output at each stage may not be symmetric.

2.7 SOFT THRESHOLD METHOD
If we enhance the high frequency co-efficient at this level image detail
information and noise all enhanced. Noise in the High Frequency Field is to be
reduced by soft Threshold method. The wavelet coefficients calculated by a
wavelet transform represent change in the time series at a particular resolution. By
looking at the time series in various resolutions it should be possible to filter out
noise. At least in theory. However, the definition of noise is a difficult one. One of
29

my colleagues commented once that "one person's noise is another's signal". In
part this depends on the resolution one is looking at. The algorithm is:

1. Calculate a wavelet transform and order the coefficients by increasing
frequency. This will result in an array containing the time series average plus a set
of coefficients of length 1, 2, 4, 8... The noise threshold will be calculated on the
highest frequency coefficient spectrum (this is the largest spectrum).
2. Calculate the median absolute deviation on the largest coefficient spectrum.
The median is calculated from the absolute value of the coefficients. The equation
for the median absolute deviation is shown below:

Here c
0
, c
1
, etc... are the coefficients.
The factor 0.6745 in the denominator rescales the numerator so that is also
a suitable estimator for the standard deviation for Gaussian white noise (Wavelet
Methods for Time Series Analysis).
3. For calculating the noise threshold I have used a modified version of the
equation in Wavelet Methods
4. Apply a thresholding algorithm to the coefficients. There are two popular
versions:
Hard thresholding. Hard thresholding sets any coefficient less than or equal to
the threshold to zero.
if (coef[i] <= thresh)
coef[i] = 0.0;
Soft thresholding. Hard thresholding sets any coefficient less than or equal to
the threshold to zero. The threshold is subtracted from any coefficient that is
greater than the threshold. This moves the time series toward zero.
if (coef[i] <= thresh)
coef[i] = 0.0;
else
coef[i] = coef[i] - thresh;

30

5. I have used hard thresholding. Soft thresholding not only smooths the time
series, but moves it toward zero. For my applications this is not a desirable
property. To find the threshold

jil jil
N T log 2 =
Where ( )
2
1 \
1

=
=
Njil
k
jil
k
jil
jil
jil
x x
N


2.8 WEIGHTED CO-EFFICIENTS

Different high frequency sub images denote the different detailed information of
an image .
Different weighted values used for different sub images
( ) ) , ( ), , ( y x WjilG W y x G M
jil
=
Where
G(x, y) = denotes denoised high-frequency coefficients
M (G(x, y), W
jil
) = denote enhanced coefficients.
W
jil
= weight coefficients
Through inverse wavelet and haar transform the enhanced image is produced

2.9 HISTOGRAM EQUALIZATION

To improve the better visualization enhanced gray scale range stretched by
non-linear histogram equalization.
This technique for equalizing the histogram
It gives the best possible dynamic range and strong contrast .
It makes the image details more visible.





31

CHAPTER 3
SOFTWARE SPECIFICATION
3.1 INTRODUCTION TOVECTOR MANIPULATIONS
In Matlab applications it is often necessary to create vectors, to delete
elements of given vectors, to substitute elements in given vectors, to append
vectors or to insert a given vector into another vector. All these methods are
discussed in this short tutorial. The corresponding m-files are available.
3.1.1 SIMPLE 3D-PLOTS
Plot a two-dimensional Gaussian probability density function f
x,y
(x, y) with zero
mean and variance 1 within the range x [5, 5], y [5, 5]:
( )
( )
2 2
2
1
,
2
1
,
y x
y x
e y x f
+
=

3.2 BASIC PLOT ROUTINES
3.2.1 2D-Routines
plot(X,Y): Plots vector Y versus vector X. If X or Y is a matrix, then the vector
is plotted versus the rows or columns of the matrix, whichever line up. If X is a
scalar and Y is a vector, length(Y) disconnected points are plotted. If X is left out,
vector Y is plotted versus its indices.
bar(X,Y): draws the columns of the M-by-N matrix Y as M groups of N vertical
bars. The vector X must be monotonically increasing or decreasing.
stairs(X,Y): draws a stairstep graph of the elements of vector Y. stairs(X,Y)
draws a stairstep graph of the elements in vector Y at the locations specified in X.
The X-values must be in ascending order and evenly spaced.
stem(X,Y): plots the data sequence Y as stems from the x-axis terminated
with circles for the data value. stem(X,Y) plots the data sequence Y at the values
specified in X.
semilogx(...), semilogy(...), loglog(...):
is the same as plot(...), except a logarithmic (base 10) scale is used for the x-axis,
y-axis or both.
3.2.2 3 D-Routines
mesh(X,Y,Z,C): plots the Coloured parametric mesh defined by four matrix
arguments. The viewpoint is specified by view. The axis labels are determined by
the range of X, Y andZ, or by the current setting of axis. The Colour scaling is

32

determined by the rangeof C, or by the current setting of caxis. The scaled Colour
values are used as indicesinto the current Colourmap.
surf(X,Y,Z,C): is the same as mesh. The shading model is set by shading.
surfl(X,Y,Z,C):is the same as surf(...) except that it draws the surface with
highlights from alight source.
contour(Z): is a contour plot of matrix Z treating the values in Z as heights
above a plane. Acontour plot are the level curves of Z for some values V. The
values V are chosenautomatically. contour(X,Y,Z): X and Y specify the (x,y)
coordinates of the surfaceas for surf.
3.2.3 Image formats
BMP
HDF
JPEG
PCX
TIFF
XWB
Most images you find on the Internet are JPEG-images which is the name for
one of the most widely used compression standards for images. If you have stored
an image you can usually see from the suffix what format it is stored in. For
example, an image named myimage.jpg is stored in the JPEG format and we will
see later on that we can load an image of this format into Matlab.
3.2.3.1 Intensity image (gray scale image)
This is the equivalent to a "gray scale image" and this is the image we will
mostly work with in this course. It represents an image as a matrix where every
element has a value corresponding to how bright/dark the pixel at the
corresponding position should be Coloured. There are two ways to represent the
number that represents the brightness of the pixel: The double class (or data
type). This assigns a floating number ("a number with decimals") between 0 and 1
to each pixel. The value 0 corresponds to black and the value 1 corresponds to
white. The other class is called uint8 which assigns an integer between 0 and 255
to represent the brightness of a pixel. The value 0 corresponds to black and 255 to
white. The class uint8 only requires roughly 1/8 of the storage compared to the
class double. On the other hand, many mathematical functions can only be applied
to the double class. We will see later how to convert between double and uint8.
33

Binary image
This image format also stores an image as a matrix but can only Colour a
pixel black or white (and nothing in between). It assigns a 0 for black and a 1 for
white.
Indexed image
This is a practical way of representing Colour images. (In this course we will
mostly work with gray scale images but once you have learned how to work with a
gray scale image you will also know the principle how to work with Colour images.)
An indexed image stores an image as two matrices. The first matrix has the same
size as the image and one number for each pixel. The second matrix is called the
Colour map and its size may be different from the image. The numbers in the first
matrix is an instruction of what number to use in the Colour map matrix.
RGB image
This is another format for Colour images. It represents an image with three
matrices of sizes matching the image format. Each matrix corresponds to one of
the Colours red, green or blue and gives an instruction of how much of each of
these Colours a certain pixel should use.
Multiframe image
In some applications we want to study a sequence of images. This is very
common in biological and medical imaging where you might study a sequence of
slices of a cell. For these cases, the multiframe format is a convenient way of
working with a sequence of images. In case you choose to work with biological
imaging later on in this course, you may use this format.

How to convert between different formats
The following table shows how to convert between the different formats
given above. All these commands require the Image processing tool box!

Image format conversion
(Within the parenthesis you type the name of the image you wish to convert.)




34

Table 3. Matlab Operation with commands
Operation: Matlab command:
Convert between intensity/indexed/RGB
format to binary format.
dither()
Convert between intensity format to
indexed format.
gray2ind()
Convert between indexed format to
intensity format.
ind2gray()
Convert between indexed format to RGB
format.
ind2rgb()
Convert a regular matrix to intensity format
by scaling.
mat2gray()
Convert between RGB format to intensity
format.
rgb2gray()
Convert between RGB format to indexed
format.
rgb2ind()

The command mat2gray is useful if you have a matrix representing an
image but the values representing the gray scale range between, let's say, 0 and
1000. The command mat2gray automatically re scales all entries so that they fall
within 0 and 255 (if you use the uint8 class) or 0 and 1 (if you use the double
class).

3.4 HOW TO READ FILES
When you encounter an image you want to work with, it is usually in form of
a file (for example, if you down load an image from the web, it is usually stored as
a JPEG-file). Once we are done processing an image, we may want to write it
back to a JPEG-file so that we can, for example, post the processed image on the
web. This is done using the imread and imwrite commands. These commands
require the Image processing tool box!
3.4.1 Loading and saving variables in Matlab
This section explains how to load and save variables in Matlab. Once you
have read a file, you probably convert it into an intensity image (a matrix) and work
35

with this matrix. Once you are done you may want to save the matrix representing
the image in order to continue to work with this matrix at another time. This is
easily done using the commands save and load. Note that save and load are
commonly used Matlab commands, and works independently of what tool boxes
that are installed.

3.4.2 How to display an image in Matlab
Here are a couple of basic Matlab commands (do not require any tool box) for
displaying an image.
Displaying an image given on matrix form
Table 4 Matlab command on Displaying an image
Operation: Matlab command:
Display an image represented as the
matrix X.
imagesc(X)
Adjust the brightness. s is a parameter
such that -1<s<0 gives a darker image,
0<s<1 gives a brighter image.
brighten(s)
Change the Colours to gray. Colourmap(gray)

Sometimes your image may not be displayed in gray scale even though you
might have converted it into a gray scale image. You can then use the command
Colourmap(gray) to "force" Matlab to use a gray scale when displaying an image.
If you are using Matlab with an Image processing tool box installed, I recommend
you to use the command imshow to display an image.
Displaying an image given on matrix form (with image processing tool box)
Table 3.10.2 (b) Matlab command on Displaying an image in matrix form
Operation: Matlab command:
Display an image represented as the matrix
X.
imshow(X)
Zoom in (using the left and right mouse
button).
zoom on
Turn off the zoom function. zoom off

36

CHAPTER 4
CONCLUSION
The method developed in this Letter is used to enhance the medical image.
We use two level wavelet transform here. In the step of the nonlinear histogram
equalisation, we set M 255/3 and N fmax/4. The enhancement weight
coefficient is 1.5. Fig. 2 shows experiment results. Figs. 2a and d are original
images, Figs. 2b and e are enhanced images via the proposed algorithm, and
Figs. 2c and f are the results of histogram equalisation. The PSNR value of Fig. 2b
is 39.64, and of Fig. 2c 30.26, the PSNR value of Fig. 2e is 70.53, and that of Fig.
2f is 45.53. From the enhanced results, enhanced images with the method
proposed are better than results with histogram equalisation. In Figures , not only
are the images blur and low-contrast detail enhanced, but the images textures
are clearer.

















Fig: 22 Wavelet Enhancement diagram




From the above image shows that the enhancement image by proposed
method in wavelet enhancement method .



37

And that the non linear histogram output which is shown in the below figure











Fig: 23 Non linear method image


















38

CHAPTER 5
RESULTS

An important problem of medical image enhancement based on wavelet
transform is how to extract high-frequency information. Haar transform is used to
decompose the high-frequency sub-images of wavelets in this Letter. This helps
us to extract high-frequency information effectively. Different enhancement weight
coefficients in different sub-images and nonlinear histogram equalisation are used
in the process of medical image enhancement. They can
also help us to enhance a medical image effectively. Results of experiments show
that the algorithm not only can enhance an images contrast, but also can preserve
the original images edge property effectively. We use two level wavelet transform
to decompose an image in this Letter. The level of wavelet decomposition is
preferably not more than four times for reasons of soft threshold filtering, else it
may reduce the quality of detailed information.

REFERENCES
1. Haar, Alfred; Zur Theorie der orthogonalen Funktionensysteme. (German)
Mathematische Annalen 69 (1910), no. 3, 331371.
2. B. Lee, Y.S. Tarng, Application of the discrete wavelet transform to the
monitoring of tool failure in end milling using the spindle motor current, The
International Journal of Advanced Manufacturing Technology, 1999, Volume
15(4), p. 238-243, ISSN 0268-3768 (Print) 1433-3015 (Online), DOI
10.1007/s001700050062.
3. Gilbert G. Walter, Xiaoping Shen, Wavelets and Other Orthogonal Systems
(2001)
4. Haar A. Zur Theorie der orthogonalen Funktionensysteme, Mathematische
Annalen, 69, pp 331371, 1910.
5. Charles K. Chui, An Introduction to Wavelets, (1992), Academic Press, San
Diego,
6. V. Buzuloiu, M. Ciuc, R. M. Rangayyan, & C. Vertan, Adaptive neighborhood
histogram equalization of color images, International Journal of Electron Image,
10(2), 2001, 445-459.
39

7. P. E. Trahanias, & A. N. Venetsanopoulos, Color image enhancement
through 3-D histogram equalization, Proc. 11th IAPR Conf. on Pattern
Recognition, The Hague, Netherlands, 1992, 545-548.
8. B. A. Thomas, R. N. Strickland, & J. J. Rodriguez Color image
enhancement using spatially adaptive saturation feedback, Proc. 4th IEEE
Conf. on Image Processing, Santa Barbara, CA, USA, 1997, 30-33.
9. A. Gupta, & B. Chanda, A hue preserving enhancement scheme for a class
of color images, Pattern Recognition Letters, 17(2), 1996, 109-114.
10. L. Lucchese, S. K. Mitra, & J. Mukherjee, A new algorithm based on
saturation and desaturation in the xychromaticity diagram for enhancement and
re-rendition of color images, Proc. 8th IEEE Conf. on Image Processing,
Thessaloniki, Greece, 2001, 1077-1080.
11. M. S. Shyu, & J. J. Leou, A genetic algorithm approach to color image
enhancement, International Journal of Pattern Recognition, 31(7), 1998, 871-
880.
12. A. Toet, Multiscale color image enhancement, Pattern Recognition Letters,
13(3), 1992, 167174.
13. J. Lu, & D. M. Hearly, Contrast enhancement via multi-scale gradient
transformation, Proc. SPIE Conf. on Wavelet
ApplicationOrlandoFLUSA1994, 345-365.
14. T. J. Brown, An adaptive strategy for wavelet based image enhancement,
Proc. IMVIP Conf. on Irish Machine Vision and Image Processing, Belfast,
Northern Ireland, 2000, 67-81.
15. R. S. Ledley, M. Buas, & T. J. Colab, Fundamentals of true-color image
processing, Proc. 10th IEEE Conf. on Pattern Recognition, Los Alamos, CA,
USA, 1990, 791-795
16. S. K. Naik, & C. A. Murthy, Hue-preserving color image enhancement
without gamut problem, IEEE Trans.on Image Processing, 12(12), 2003, 1591-
1598.
17. C. C. Yang, & J. J. Rodrguez, Efficient luminance and saturation
processing techniques for bypassing color coordinate transformations, Proc.
IEEE Conf. On Systems, Man and Cybernetics, Bogota, Columbia, 1995, 56-67.
40

18. Y. Xu, J. B. Weaver, D. M. Healy, & J. Lu, Wavelet transform domain
filters: A spatially selective noise filtration technique, IEEE Trans. on Image
Processing, 3(6), 1994, 747-758.
19. Q. Pan, L. Zhang, G. Dai, & H. Zhang, Two denoising methods by wavelet
transform, IEEE Trans. On Signal Processing, 47(12), 1999, 3401-3406.
20. Y. Kobayashi, & T. Kato, A high fidelity contrast improving model based
on human vision mechanism,Proc. IEEE International Conf. on Multimedia
Computing and Systems, Florence, Italy, 1999, 578-584.
21. R. N. Strickland, C. S. Kim, & W. F. McDonnell, Digital color image
enhancement based on the saturation component, International Journal of
Optical Engineering,26(7), 1987, 609-616.
22. Jobson D. J., Rahman Z., & Woodell G. A., Statistics of visual
representation, Proc. SPIE Conf. on Visual nformation Processing XI, Orlando,
FL, USA, 2002, 25-35