International Journal of Advances in Electronics and Computer Science, ISSN: 2393-2835
Volume-3, Issue-5, May.-2016
STUDY OF COMPRESSION TECHNIQUE EZW, SPIHT AND
3D_SPIHT
TANUJA DAVE
Computer science Engineering, Department computer science, India
Shri Vaishanav Institute of technology and science, Indore
E-mail: [email protected]
Abstract— In this paper, we study the three compression technique. i.e., Embedded Zero tree wavelet (EZW), Set
Partitioning in Hierarchical Trees (SPIHT), 3D_SPIHT.there are three types of Wavelet transform are applied on the image.
Compressed image is achieved by using the above compression technique and from the compressed image, decompressed
image can be retrieved. The quality of the decompressed image is find out using the parameter i.e. Mean Square Error
(MSE), Peak Signal to noise Ratio (PSNR) and Maximum Difference (MD). After analyzing this parameter we examine
which technique is better for image compression.
Keywords— EZW, SPIHT, 3D_SPIHT.
discrete wavelet transform (DWT) [4], which helps us
to determine the order of importance. In addition,
arithmetic coding is used to achieve a fast and
effective method for entropy coding.
I. INTRODUCTION
Image compression is minimizing the size of a
graphic file without degrading the image quality of
image. The reduction in file size allow more image to
be stored in a given amount of memory space. images
are widely used coding algorithms based on wavelet
transform i.e. embedded zero-tree wavelet (EZW)
algorithm , 3D_SPIHT and the set partitioning in
hierarchical trees (SPIHT) algorithm.
2.2. SPIHT
SPIHT is an embedded wavelet coding algorithm.
The SPIHT algorithm is the next version of the EZW
algorithm. It was devised by Said and Pearlman [4,
5]. Some of the best results i.e. highest PSNR values
are obtained with SPIHT.
The SPHT algorithm is different because it does not
directly transmit the contents of the sets the pixel
coordinates [6] but it send decisions which is made in
every node of the trees that define the structure of the
image. Because here only decisions are being
transmitted, the pixel value is defined by in which
points the decisions are taken and their outcomes,
while the coordinates of the pixels are defined, which
tree and what part of the tree, the decision is being
made on. The advantage to this is that the decoder
have an identical algorithm which is able to identify
with each of the decisions and can create identical
sets along with the encoder.
Fig 1: Image compression model using wavelet transform
Fig 2: Image decompression model using wavelet transform
Image compression and decompression model of
wavelet is given above in the fig 1 and fig 2.
II. COMPRESSION TECHNIQUES
2.3. 3D_ SPIHT
Exploitation the self-similarity across spatial
temporal orientation tress.in this way the compressed
bit stream will be fully embedded, so that a single
video file sequence can provide efficient video
quality, this algorithm can run any compressed file
size and let it run until nearly lossless reconstruction
is obtained, which is required in many application
including HDTV [5].
2.1. EZW
The embedded zero tree wavelet algorithm (EZW) [1]
is the image compression algorithm. This method
produces embedded bit stream for image coding.
This technique requires neither training nor pre-stored
codebooks or tables and requires no preceding
knowledge of the source image. The EZW algorithm
is based on four principal i.e. 1) a discrete wavelet
transform or decomposition of hierarchical sub-band
2) prediction of absence of significant information
III. EXPERIMENTS
3)
Entropy
coded
successive-approximation
quantization and 4) universal lossless data
compression which is accomplished using adaptive
Arithmetic coding. The EZW algorithm includes
Below are the four figure first is original image and
the rest three images are the decompressed image
using the above three technique.
Study of Compression Technique EZW, SPIHT and 3D_SPIHT
51
International Journal of Advances in Electronics and Computer Science, ISSN: 2393-2835
Volume-3, Issue-5, May.-2016
Table 2:
As discussed from the above table size are same as
the original input image but the quality are not same
as the original image. The quality of the image is find
by performance parameter.
\
Fig 3: original image
IV. PERFORMANCE ANALYSES
These three techniques are implemented and the
results are shown in the table 1 and table 2. The
PSNR, MSE and MD values are calculated in Table
3.
4.1. Peak Signal to Noise Ratio (PSNR in dB)
The PSNR can give us the quality of image, but itself
it cannot make a comparison between the qualities of
two different image. It is possible that an image with
a lower PSNR have the better quality compared to
one with a greater signal to noise ratio.
Fig 4: decompressed with EZW
Peak Signal-to-Noise Ratio is an engineering term.
To find the ratio between the maximum possible
power of a signal and corrupting noise. Many signals
have a very large dynamic range, PSNR is expressed
in terms of the logarithmic decibel scale.
Fig 5: decompressed with SPIHT
4.2. Mean Square Error (MSE)
Mean square error is a estimator measure of image
quality index. The large value of mean means that
image is a low quality. It represents the classical error
estimate given by the equation:
4.3. Maximum Difference (MD)
It is the Difference between two pixels such that the
larger pixel comes after the smallest pixel. If the
maximum difference value is high , means image is
poor in quality.
Fig 6: decompressed with 3D_SPIHT
Table 1:
After compression of the image, do the
decompression by taking the compressed image as an
input image and find the size of image, which is
shown in table 2.
Study of Compression Technique EZW, SPIHT and 3D_SPIHT
52
International Journal of Advances in Electronics and Computer Science, ISSN: 2393-2835
many images. The experimental results shows that the
spiht_3d technique performs better when we want
less MSE and MD and SPIHT is better for getting the
less PSNR.
V. COMPRESSION RATIOS

For EZW
CR = (1−7.02/192) ∗100
= 96.34%

For SPIHT
Volume-3, Issue-5, May.-2016
REFERENCES
[1]

CR = (1−4.97/192) ∗100
=97.414%
For 3D_SPIHT
[2]
CR = (1−4.37/192) ∗100
=97.7239%
[3]
CONCLUSIONS
[4]
In this paper, the results are compared using three
different
wavelet-based
image
compression
techniques. Compression ratio is calculated. The
results of the above three techniques ‘EZW’, SPHIT
and Spiht_3d are compared by taking the
[5]
[6]
Parameters such as PSNR, MSE and MD values from
the input image. These techniques are applied in
J. Tian and R.O. Wells, Jr. “A lossy image codec based on
index coding” IEEE Data Compression Conference, DCC
’96, page 456, 1996.
J. Tian and R.O. Wells, Jr. “Embedded image coding using
wavelet difference reduction” ,Wavelet Image and Video
Compression, P. Topiwala, ed., pp. 289–301. Kluwer
Academic Publ., Norwell, MA, 1998.
A. Said and W.A. Pearlman.” A new, fast, and efficient
image codec based on set partitioning in hierarchical trees”
IEEE Trans. on Circuits and Systems for Video Technology,
Vol. 6, No. 3, pp. 243–250, 1996
J. M. Shapiro, “Embedded image coding using zero trees of
wavelet coefficients” IEEE
Transactions on Signal
Processing, vol. 41, No. 12,pp. 3445-3462, 1993
Beong Jo Kim, Zixiang Xiong,William a. Pearlman," Low
bit Rate, Scalable Video coding with 3D Set Partitioning in
Hierarchical Trees" IEEE Trans. on Circuits and Systems for
Video Technology, Vol. 10, No. 8, pp. 1374–1387, 2000..
Kassim A Lee W S. “Color image coding using SPIHT
with partially linked spatial orientation trees “. IEEE Trans
on Circuit and System for Video Technology, Vol.2, No 2,
203-206, 2003.

Study of Compression Technique EZW, SPIHT and 3D_SPIHT
53