Breast and with the help of mini-MIAS database, results

Breast cancer becoming the most occurring forms of
cancer in women, especially in developing countries as well as in developed
countries. Mammogram is accepted as best radiographic method to clinical
examination in the early stages of illnesses for breast cancer. Diagnostic
features such as cysts and masses may be small or ill-defined, so mammogram
image quality is extremely important when used for interpretation and
diagnosis. Pre-processing is an important
part of any imaging modalities to bring the image to that quality where it is
suitable for analysis and extraction of significant data. This
paper talks about, six denoising techniques which have great
significance in mammographic image analysis due to poor quality of mammograms and
with the help of mini-MIAS database, results were evaluated using many image metrics.  Cancer
is abnormal change, in the genes which are responsible for regulating the
growth of cells and keeping them healthy, where genes in each cell’s nucleus,
which acts as the “control room” of each cell. Normally, the process of cell
growth like old cells in our bodies replaces themselves through a healthy new. Sometimes
changed cell gains the ability to keep dividing without control or order, producing
more and forming a tumor 1. Breast cancer is one of the most occurring forms
of cancer in women, especially in India. Breast Cancer detection at initial
stage is the only solution, to reduce the death rate among the women’s, so
regular screening of mammograms for detection of masses is essential, so wide range of
treatment options can be available at early stages  like less-aggressive surgery and adjuvant
therapy. 2Mammography is a medical imaging which uses a low-dose x-ray
system to analyze inside the breasts. A mammography examination is known
as mammogram, which
aids in the early detection and diagnosis of breast diseases in women. Mammogram is accepted as an important adjunct to clinical
examination in the care of patients with many common illnesses for breast
cancer.  It provides
the complete information about shape, structure and pathologies of breast which
helps out in diagnosing breast cancer.  Mostly mammograms are found with five abnormalities which
require further examination to classify it as benign or malign. The
mammograms are confirmed as abnormal after the clinical diagnosis mammography which
is divided into two classes: benign (noncancerous) and malignant (cancerous). The abnormalities are Asymmetric
breast tissue, Mass, Micro calcification, Asymmetric density and Architectural
distortion. Except for mass and micro calcification all the other abnormalities
are found in very rare cases known as invasive breast cancer.3 Diagnostic features such as a benign or
malignant tumor, or non-masses are exclusively normal tissues which may be small or ill-defined, so
mammogram image quality is extremely important when used for interpretation and
diagnosis.4In digital imaging, systems introduce various noises and artifacts which
are to be removed. The noise like
low contrast of the small tumors to the background can be hardly detected in
mammography. For this an image preprocessing techniques are adopted to reduce
the noise level of the image, preserving the mammography structures and
features. Also denoising is more significant in image processing, analysis and applications;
by reserving the details of an image and removing the random noise as far as
possible is the goal of image denoising approaches. In this work, a comparative study of several
denoising techniques for mammogram is presented. The filters considered
are:  1) median filter, 2) adaptive
median filter, 3) frost filter, 4) wavelet filter, 5) histogram equalization,
6) Contrast Limited Adaptive Histogram Equalization.A traditional way to
remove noise from image data is to employ spatial filters. In Spatial
filtering the processed value for the current pixel depends on both itself and
surrounding pixels. Spatial filter are classified into non-linear and linear filters.Linear
filters process time-varying input signals to produce output signals, subject
to constraint of linearity whereas non-linear filters produces output signals, subject
to non-linearity especially in removal of certain types of noise that are not
additive.Median filtering is a nonlinear process which is useful in
preserving edges in an image and also in reducing noise.
In this filter each pixel in image looks at its adjacent neighbor pixel to
decide whether or not it is representative of its surroundings and replace the
pixel value with the median of those values. In median filtering
,sorting all the pixel value within the window size from the surrounding
neighbor into numerical order and then replace the pixel with median intensity
value of the pixels within the windows. Adaptive median filter5 is an improved
version of median filter which works within a rectangular region(window) with
each output pixel contains the median value of window neighborhood around the
corresponding pixel in the input images. This filter is basically used to
smooth the non-repulsive noise from two-dimensional signals without blurring
edges and preserved images which makes, it suitable for enhancing mammogram
images. Frost filter 7 is an exponentially
weighted averaging filter which is used to eliminate the quantum noise
from the mammograms in which the coefficient of
variation is the ratio of the local standard deviation to the local mean of the
distorted image. Frost filter uses weighted sum of values with ‘n’ window size
to replace central pixel and this weighted factor directly proportional to the
difference between central and other pixels.Wavelet transforms filtering have become increasingly important which
play an extremely crucial role in image processing since it allow both time and
frequency analysis simultaneously. Wavelet transform
decomposes the input image into four lower sub-bands with approximation
coefficients (LL1) and detail coefficients (HH1, HL1, and LH1). To obtain the
next level of wavelet coefficients, the sub-band LL1 is further decomposed and
sampled which results in two-level wavelet decomposition. In this technique
,elimination of any of the undesired sub-band or their combinations are done
and then reconstructing the original image 
using inverse wavelet transform. In our experiments, we use Haar wavelet
and eliminate HH, LH, and HL bands individually for first- and second-level

Histogram equalization is a nonlinear contrast
enhancement which is basically used for enhancing the appearance of images. Histogram
equalization enhances the contrast of images by transforming the values in an
intensity image, or an indexed image, in such a manner that the output image
histogram approximately matches a given histogram.( give a linear
trend to the cumulative probability function associated to the image.)For images which contain local
regions of low contrast bright or dark regions, histogram equalization doesn’t
work effectively so a modified histogram equalization technique called Contrast Limited Adaptive Histogram Equalization can be used on such images
for better results which consider only small regions and based on their local cumulative distribution function(cdf), performs contrast
enhancement of those regions.6 The amount of contrast enhancement for some
intensity is directly proportional to the slope of the CDF function at that
intensity level. Hence contrast enhancement can be limited by limiting the
slope of the CDF. The
slope of CDF at a bin location is determined by the height of the histogram for
that bin.
Therefore if we limit the height of the histogram to a certain level we can
limit the slope of the CDF and hence the amount of contrast enhancement. The CLAHE method
seeks to reduce the noise and edges-shadowing effect produced in homogeneous
areas and was originally developed for medical imaging. This method has been
used for enhancement to remove the noise and reduces the edge-shadowing effect
in the pre-processing of digital mammogram.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Measurement of image quality is very important to many
image processing systems. Since it inherent physical limitations and economic
reasons for the quality of images and videos, such that it could view by a
human observer. Some of existing measures of image quality are listed below.Image
differencing is used to
determine changes between images and Maximum difference
(MD) is the maximum of the error between real image and filtered image. For
the better performance maximum difference should be minimum and large value
shows poor quality.The root mean square error (RMSE) is the
square root of cumulative squared error between the filtered and the original
image for comparing various image processing. For the better performance
root mean square error should
be minimum and large value shows poor quality.The peak signal-to-noise ratio(PSNR) in image processing is a
standard measure of the sensitivity of imaging system. For the better performance peak signal-to-noise ratio should
be maximum and smaller value shows poor quality.In image processing normalized
absolute error (NAE) is a measure
of difference between the filtered and the
original image for comparing various images. Higher value of normalized absolute
error is used for poor quality of images.Smaller the value of structural
content (SC) better is the image quality and higher value of structural
content is used for poor quality of images. For image-processing
applications in which the brightness of the image and template can vary due to
lighting and exposure conditions, the images can be first normalized. Normalized Cross Correlation (NCC) show
better quality with higher value. Laplacian
mean squared error (LMSE) is based on the importance of edges measurement
with lower the value of laplacian mean squared error better the quality of
images. Universal image quality index (IQI) is a image quality
distortion measurement which is defined as the product of three factors:
structural distortion, contrast distortion and luminance distortion, of the
distortion of ideal image with respect to filtered image.The Structural Similarity (SSIM) index
is a method for measure similarity between ideal image and filtered image. It
is an quality measure of one of the filtered images being compared, provided the ideal image is regarded as of
perfect quality . SSIM is a better than that of universal
image quality index. The Pratt’s Figure of Merit(PRATT)
find the edge location accuracy by the displacement of filtered image from an
ideal image where both the image (filtered & ideal) are edge detected
images from the edge detectors like sobel , prewitt , Robert , log etc.

The Mammographic Image Analysis Society (MIAS)
database is the largest publicly available database of mammographic data. It contains approximately 322 screening mammography cases,
where 207 images represent normal, while 64 and 51 images referred as benign
and malignant cases respectively.For experimental purpose, the 45 images
are taken from MIAS database which includes of 15 normal images and 30 abnormal
images. The abnormal images are again classified into two classes which are
benign and malign. There are 15 benign images and 15 malign images.In this section, the results of those filters are compared
which are discussed in previous section. Many proposed work in the literature
are also discussed and compared with the present approach. Furthermore, the
filtered images shown in this section are obtained as a result of subjecting
the breast mammogram shown in Figure1 to
various filters.The image quality metrics of median filter are presented in Table 1 and the results of median
filter applied to the original mammogram of Figure1 are shown in Figure 2 . From the image quality metrics table it is found that, the
image quality degrades with the increase in windows size. Some of the
performance of image metrics (MD,NAE,LMSE,RMSE,SC) increase and some (PSNR,IQI,
SSMI,NCC) decreases with the increase of windows size .Also pratts figure of
merit show better performance with lower window size.Table 2
shows the image quality metrics obtained for Adaptive median filter (AMF) and Figure
3 shows the results for AMF. In AMF the image quality will not change much with
the increase in windows size .It is found that performance of AMF is quite good
resulting in lower value of RMSE and higher value of PSNR. The Pratts figure of merit also shows better results as compared to
median filter.Figure 4 shows the result of frost filter and the image quality
metrics obtained for frost filter is presented in Table 3. Here the image quality will change variably with the
increase in windows size .It is found that performance of frost filter is quite
good resulting in window size 5.Figure 5 shows the filtered images from
wavelet filtering, and the corresponding tables of image quality metrics are
presented in Table 4. It is found that image quality
is maintained after filtering at first-level decomposition as indicated by
RMSE, PSNR, IQI and SSMI while after second-level decomposition, image becomes
much brighter and Pratt’s figure of merit reduces. When LH band is eliminated
after second-level decomposition, most of the details are lost giving MD of
0.000 and NAE of 254.00 .The best result is obtained when HH band is eliminated
after first-level decomposition.The image quality metrics obtained for histogram
equalization(column 2) & CLAHE(column 3) is presented in Table 5 and the results for histogram equalization (a)& CLAHE (b)are
shown in figure 6. It is found that with maximum difference and psnr in
histogram equalization is smaller than all other filters.In this paper, review and comparison of representative denoising
methods both qualitatively and quantitatively with extensive experiments
conduct to evaluate the performance of all the algorithms. In analytical
comparison, it was found that image representations with over complete basis
functions improve the performance within each category.In this paper it is
clear from the comparison that all the denoising techniques are important for
various applications. In applications that require high efficiency, some
filters are used, some filters are more appropriate for high searching
complexity,  memory and complexity issue