COMPARATIVE ANALYSIS BETWEEN DISCRETE COSINE

TRANSFORM AND WAVELET TRANSFORM TECHNIQUES FOR

MEDICAL IMAGE COMPRESSION.
Ajala Funmilola A*, Fenwa Olusayo D*, Aku Michael A.,
Department of Computer Science and Engineering, LAUTECH Ogbomoso, Oyo state, Nigeria.

*E-mail of the corresponding author: funfaith2003@yahoo.co.uk. fenwadeborah@yahoo.com.

ABSTRACT
Image compression reduces irrelevant and redundancy of the image data in order to be able to store or
transmits data in an efficient form. Image compression is minimizing the size in bytes of a graphics file
without degrading the quality of the image to an unacceptable level. The reduction in file size allows
more images to be stored in a given amount of disk or memory space. It also reduces the time required
for images to be sent over the Internet or downloaded from Web pages. Medical image compression
plays a key role as hospitals move towards filmless imaging and completely digital. Image compression
will allow Picture Archiving and Communication Systems (PACS) to reduce the file sizes on their
storage requirements while maintaining relevant diagnostic information. Teleradiology sites benefit
since reduced image file sizes yield reduced transmission times. Even as the capacity of storage media
continues to increase, it is expected that the volume of uncompressed data produced by hospitals will
exceed capacity and drive up costs. The improved compression performance will be accomplished by
making use of clinically relevant regions as defined by physicians. This work compared Discrete Cosine
Transform (DCT) compression technique and Wavelet Transform (WT) compression techniques for
medical images. The result showed compression ratio of 10:1 and 7:1 for DCT and WT respectively.
The mean difference of 77.84 with standard deviation of 83.17 and mean difference of 77.77 with
standard deviation of 83.23 from the original image were recorded for DCT and WT compression
technique.

1. INTRODUCTION

There are two forms in which medical image compression can be done which are namely; Lossless
compression (compression done involving no loss of data) and Lossy compression (compression done
involving loss of data). Lossless compression will be applied in the clinically relevant areas and lossy
compression will be applied in the other areas. Lossless compression is preferred for archival purposes
and often for medical imaging, technical drawings, clip art, or comics. This is because lossy
compression methods, especially when used at low bit rates, introduce compression artifacts. Most
doctors would prefer using lossy compression with the quality level on high. This would yield an image
that is much smaller than it’s lossless-compressed counterpart that make the image visually lossless.
However, the lossy compressed image, when decompressed, is not identical to the original i.e., there was
loss and there is a problem here.
CT or MRI Medical imaging produces human body pictures in digital form. Since these imaging
techniques produce prohibitive amounts of data, compression is necessary for storage and
communication purposes. Many current compression schemes provide a very high compression rate but
with considerable loss of quality. On the other hand, in some areas in medicine, it may be sufficient to
maintain high image quality only in the region of interest, i.e., in diagnostically important regions.

978-1-4799-7186-2/15/$31.00 ©2015 IEEE

Compression is not just about the storage costs, it is also about transmission time, imaging apparatus
utilization, convenience and comfort of the patient.

2 LITERATURE REVIEW

Many hospitals have satellite centres or clinics in small towns and remote areas to make it
convenient for patients who have a hard time travelling the distance to the hospital, especially for
diagnostic procedures. These hospitals make use of ‘teleradiology’ applications that allow the clinic staff
to operate the clinic without the need for a radiologist to be present. Instead of a diagnostic radiologist, a
technician or basic radiologist in the clinic can take the X-ray and send the image through a network
connection to the hospital where the diagnostic radiologist can read the image and send back a
diagnosis. But there is a problem, especially in emergency situations where time is of the essence,
because a 10 MB image will take approximately half an hour using a high-speed modem. (Krishnan et.
al., 2006)
Images can be compressed using lossy or lossless techniques. Lossless techniques allow the image
to be compressed, then decompressed back to the original state of the image without any loss of data
(Kivijarvi et al., 1999). These methods are sometimes called reversible compression methods.
Compression rates for lossless techniques vary but typically are around 2:1 to 3:1. The most popular
compression algorithms in use today in the medical community are lossless JPEG (Joint Photographic
Experts Group) (Ali et al, 2008) and lossless Wavelet.JPEG has been adopted by the Digital Imaging
and Communications in Medicine(DICOM) group in their widely adopted DICOM image file format,
but the wavelet compression algorithm is gaining ground. In fact, the DICOM Working Group added
support for the JPEG 2000 standard into the DICOM format in November of 2001. There have been
numerous compression research studies examining the use of compression as applied to medical images.
Zukoski et al. (2006) showed that the wavelet transform is a promising tool for image
compression providing high rates of compression while maintaining good image quality. In 1999, a
quantitative comparison of three lossy compression methods (one wavelet and two JPEG) as applied to a
variety of 12-bit medical images in conjunction with the Department of Radiology at the Hershey
Medical Center was performed.
In the most relevant work, (Cosman et al, 1997) developed a region based coding approach. They
discussed two approaches: one uses different compression methods in each region such as ‘contour-
texture’ coding and sub band decomposition coding, and the other uses the same compression method in
each region such as the discrete cosine transform but with varying compression quality in each region
used different quantisation tables

3. METHODOLOGY
The two coding techniques that are being compared are the Discrete Cosine Transform (DCT) and
Discrete Wavelet Transform (DWT). The DCT–based encoder worked by segmentation of the image
into 8*8 blocks.

A. THE DISCRETE COSINE TRANSFORM (DCT)

DCT has been successfully employed as the first step in several coding system, because it exhibits
favorable energy distribution in the frequency domain.
DCT attempts to decorrelate the image data and after, each transform coefficient can be encoded without
dropping off compression efficiency. . The DCT for an N×N input sequence can be defined as where
x=0, 1, .., n-1, in the list of length n given by:

978-1-4799-7186-2/15/$31.00 ©2015 IEEE

Where,

For u= 0, 1, 2, N-1.

N is the size of the block that the DCT is applied on. The equation calculates one entry (i, jth) of
the transformed image from the pixel values of the original image matrix. M(x, y) is the original data of
size x* y. Discrete cosine transform (DCT) is widely used in image processing, especially for
compression algorithm for encoding and decoding in DCT technique is shown below.

ALGORITHM A1 (DCT Encoding System)

Step1. Load image

Step2. The image is broken into N*N blocks of pixels. Here N may be 4, 8, 16, e t c.
Step3. Working from left to right, top to bottom, the DCT is applied to each block.
Step4. Each block’s elements are compressed through quantization means dividing by some
specific value.
Step5. The array of compressed blocks that constitute the image is stored in a drastically reduced
amount of space.

The whole image is divided into small N*N blocks then DCT is applied on these blocks. For reducing
the storage space, DCT coefficients are quantized through dividing by some value or by quantization
matrix; so that large value become small and it needs small size of space.

ALGORITHM A2 (DCT Decoding System)

Step1. Load compressed image from disk

Step2. Image is broken into N*N blocks of pixels.
Step3. Each block is de-quantized by applying reverse process of quantization.
Step4. Now apply inverse DCT on each block. And combine these blocks into an image which is
identical to the original image.

In the decoding process, N’s value was kept the same as it used in encoding process. Then de-
quantization process was performed by multiplying with quantization value or quantization matrix. As
earlier said that this is lossy technique so output image is not exact copy of original image but it
resembled the original image.

B. DISCRETE WAVELET TRANSFORM( DWT)

There are two types of wavelet in used. First one is wavelet transform and second one is discrete wavelet
transform. Wavelet analysis is computed by filter bank.

ALGORITHM B1 (WT Encoding System)

Step1. First original image have to been passed through high pass filter and low pass filter by
applying filter on each row.
Step2. Output of the both filtered images are combined into T1
978-1-4799-7186-2/15/$31.00 ©2015 IEEE
Step3. T1 is down sampled by 2.
Step4. Again, T1 has been passed through high pass filter and low filter by applying on each
column.
Step5. Output of the step4 is passed through high pass filter and low pass filter again and then
combined into T3
Step6. Lastly, down sample T3 by 2.This is our compressed image.

ALGORITHM B2 (WT Decoding System)

Step1. Extract low pass filter image and high pass filter image from compressed image
Step2. Both images are up sampled by 2.
Step3. Now we take the summation of both images into one image called R1.
Step4. Again extract low pass filter image and high pass filter image by simply dividing
vertically.
First half is low pass filtered image. And second half is high pass filter image.
Step5. Take summation of both images. That is our reconstructed image.

Though in DWT, very high compression ratio was gotten, and minimum amount of information was
lost, but, more than one level was done, then more compression ratio was gotten and the reconstructed
image is not identical to original image.

4. RESULTS AND DISCUSSION

Few medical images were used in analyzing both methods which are Discrete Cosine Transform and
Wavelet Transform. Figure 4.2 showing interface for the loading and compression process of a Medical
Resonance Image using the Discrete Cosine Technique or Wavelet Transform. Figure 4.3 (a) shows
original Images and (b) decompressed Images in DCT Compression. Figure 4.4 (a) shows reconstructed image
after DCT Compression and (b) histogram equalization of DCT Compression.

The image produced is almost identical to the original image. Figure 4.5 (a) shows original Images and (b)
decompressed Images in DWT Compression. Figure 4.6 (a) shows reconstructed image after DWT Compression
and (b) histogram equalization of DWT Compression.

Figure 4.2: Loading of MRI Image to be Compressed Using DCT or DWT Technique.

978-1-4799-7186-2/15/$31.00 ©2015 IEEE

 (a) (b)

Figure 4.3: Showing (a) Original Images (b) Decompressed Images in DCT Compression

(a) (b)
Figure 4.4 (a) Reconstructed Image after DCT Compression (b) Histogram Equalization of DCT Compression

(a) (b)

Figure 4.5: Showing (a) Original Image (b) Decompressed Image in Wavelet Transform

(a) (b)
Figure 4.6: (a) Reconstructed Image after Wavelet Transform Compression (b) Histogram Equalization of
Wavelet Transform Compression

978-1-4799-7186-2/15/$31.00 ©2015 IEEE

Mean Squared Error (MSE) is defined as the square of differences in the pixel values between the
corresponding pixels of the two images. for DCT based image compression as the window size increases
MSE increases proportionately whereas for DWT based image compression. MSE first decreases with
increase in window size and then starts to increase slowly with finally attaining a constant value..
Compression increases with increase in window size for DCT and decreases with increase in window
size for DWT.

TABLE 1: Analysis of Results between DCT and DWT.

Parameters DCT DWT

Area 110538.75 110416.25

Mean 77.8422 77.7733
Difference
Standard 83.1728 83.2156
Deviation
Correlation 0.9635 0.96448
Compression 10:1 7:1
ratio
Time (sec) 817.5792 24.3317

5. Conclusion.

DCT performs efficiently at medium bit rates. The disadvantage with DCT is that only spatial
correlation of the pixels inside the single 2-D block is considered and the correlation from the pixels of
the neighboring blocks is neglected. Blocks cannot be decorrelated at their boundaries using DCT.
DWT provides high quality compression at low bit rates. The use of larger DWT basis functions or
wavelet filters produces blurring near edges in images. DWT performs better than DCT in the context
that it avoids blocking artifacts which degrade reconstructed images.
Moreover, if Discrete Cosine Transform is widely adapted and robust method used for compression of
digital image as it has the ability to carry the most of the information in smallest number of pixels
compared to other method, the Wavelet based Transform provides better result as far as properties like
RMS error, image intensity and execution time is concerned. So Wavelet based Transform is widely
used.
References
Ali A.F, Abir J. H, Paulo L, and Dhiya A, (2008). “An adaptive hybrid image compression method and
its application to Medical images.” IEEE
Cosman P. C., Gray R.M., Olshen R.A.( 1997): Evaluating quality of compressed medical images: SNR,
subjective rating, and diagnostic accuracy, Proceedings of the IEEE, vol.82, no.6, p.919-932,.
Krishnan K. Marcellin M., Bilgin A. and Nadar M.( 2006), “Efficient Transmission of Compressed Data
for Remote Volume Visualization,” IEEE Trans. on Medical Imaging, vol. 25, no. 9, pp. 1189-1199,
September.
Kivijarvi, K, Kadambe, S., Murray, R., Boudreaux – Bartels, G. F. (1999), “Wavelet Transform Based
QRS Detector”, IEEE Trans. on Biomedical Engineering, Vol. 46, no. 7, , pp. 838-848
Zukoski, M.J., Boult, T. and Iyriboz, T. (2006) ‘A novel approach to medical image compression’, Int.
J. Bioinformatics Research and Applications, Vol. 2, No. 1, pp.89–103.
978-1-4799-7186-2/15/$31.00 ©2015 IEEE