International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013
Content based image compression and
enlargement for arbitrary resolution
Display devices.
Ms. Suma S.chavadi1, Mr.Basavaraj.d2
E&C Dept
KLS’s VDRIT, Haliyal, Karnataka, India.
ABSTRACT:-Image compression is a technique used
to reduce the storage required to save an image. The
existing image coding methods cannot support content
based image compression and enlargement for arbitrary
resolution
display
devices.
In
multimedia
communications, image retargeting is generally required
at the user end. The work presented in this paper
addresses the increasing demand of visual signal delivery
to terminals with arbitrary resolutions, without heavy
computational burden to the receiving end. In this paper,
the principle of seam carving and wavelet based SPIHT
codec are used. for each input image ,block based seam
energy map is generated at the encoder side and the
multilevel discrete wavelet transform is performed. at
the decoder side, the end user has the ultimate choice for
the spatial scalability without the need to examine the
visual content an image with arbitrary aspect ratio can
be reconstructed in a content-aware manner based upon
the side information of the seam energy map. The final
result show that, for the end users, the received images
with an arbitrary resolution preserve important and
sensitive content while achieving high coding efficiency
for transmission.
Keywords: SPIHT, DWT,Seam Carving
I.INTRODUCTION
During the past years, Image compression has been
extensively studied. Among the existing codecs, the
most commonly used ones are mainly based upon
discrete wavelet transform, such as JPEG 2000 and
SPIHT. Generally speaking, such codec’s support
excellent quality scalability and achieve high coding
efficiency in terms of rate-distortion (R-D)
performance. Furthermore, JPEG 2000 with a region
of interest (ROI).allows the ROIs of the image to be
coded with better quality than that of non-ROIs.
However, the spatial scalability is not fully supported
in these codec’s, and only dyadic resizing can be
achieved. Without considering the image content after
jpeg2000 technique introduces a new technique that is
in painting. In painting is the process of reconstructing
lost or deteriorated parts of images and videos. For
ISSN: 2231-5381
instance, in the case of a valuable painting, this task
would be carried out by a skilled image restoration
artist. In the digital world, in painting (also known as
image interpolation or video interpolation) refers to
the application of sophisticated algorithms to replace
lost or corrupted parts of the image data (mainly small
regions or to remove little defects).but in this
technique the complexity is high at the receiving end.
a number of pruning-based coding schemes have been
proposed towards visual quality rather than pixel-wise
fidelity in the work of a line-based pruning is
performed for an original video, and the pruned video
sequence is encoded using H.264/AVC; at the
receiving end, after H.264/AVC decoding, the
resultant smaller-size video frames are interpolated
back to their original size by a high-order interpolation
method. Based on the recorded results in, these
pruning-based approaches can achieve high R-D
performance; however, the spatial scalability is still
not fully supported, and furthermore, the in painting or
interpolation at the decoder side is with heavy
computation. Nowadays, as the size of portable
devices (e.g., laptops, PDAs, and mobile phones)
continue diversifying, the existing coding schemes
cannot be directly applied, and an additional image
retargeting process (e.g., down-sampling, cropping,
warping or seam carving (SC) ]) is needed in the
receiving end. However, for the resource-limited
mobile devices, it is not always possible and
economical to perform sophisticated content-aware
image resizing. Therefore, content-based spatialscalable image compression for arbitrary resolution is
becoming one of the emergent challenges for
universal access,i.e., one can access any information
over any network from anywhere through any type of
display devices. While preserve important and
sensitive image content we present two techniques
that is seam carving and SPIHT. the advantages of SC
(i.e., content-aware image resizing) and wavelet-based
SPIHT coding (i.e., high R-D performance) are well
combined and a novel content-based spatial-scalable
http://www.ijettjournal.org
Page 2270
International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013
compression scheme is proposed. Different from the
schemes in the original image is considered as a whole
and not divided into two components (ROI and nonROI) while SC is performed and the resultant seam
energy map is used to guide the scanning and
encoding order of the DWT coefficients. The SPIHTcoded bit stream and the side information of the
resultant seams are transmitted to the decoder side. In
this way, we can reconstruct the content-aware image
with arbitrary aspect ratio. Effective resizing of
images should not only use geometric constraints, but
consider the image content as well. We present a
simple image operator called seam carving that
supports content-aware image resizing for both
reduction and expansion. A seam is an optimal 8connected path of pixels on a single image from top to
bottom, or left to right, where optimality is defined by
an image energy function. By repeatedly carving out
or inserting seams in one direction we can change the
aspect ratio of an image. By applying these operators
in both directions we can retarget the image to a new
size. The selection and order of seams protect the
content of the image, as defined by the energy
function. Seam carving can also be used for image
content enhancement and object removal. We support
various visual saliency measures for defining the
energy of an image, and can also include user input to
guide the process. By storing the order of seams in an
image we create multi-size images that are able to
continuously change in real time to fit a given size.
II.BACKGROND REVIEW
We present a simple image operator called seam
carving that supports content-aware image resizing for
both reduction and expansion. A seam is an optimal 8connected path of pixels on a single image from top to
bottom, or left to right, where optimality is defined by
an image energy function. By repeatedly carving out
or inserting seams in one direction we can change the
aspect ratio of an image. By applying these operators
in both directions we can retarget the image to a new
size. The selection and order of seams protect the
content of the image, as defined by the energy
function. Seam carving can also be used for image
content enhancement and object removal. We support
various visual saliency measures for defining the
energy of an image, and can also include user input to
guide the process. By storing the order of seams in an
image we create multi-size images that are able to
continuously change in real time to fit a given size. let
si denote the ‘I’th component of a seam, and then six
and siy represent its vertical and horizontal aspects,
respectively .mathematically, a vertical seam is
defined as
Sx={Six}i=1N={x(i),i}i=1N |x(i)-x(i-1)≤1|
ISSN: 2231-5381
Where x is a mapping of
x:[1………, N]→[1…….,M],
Similarly, If y is a mapping of
y: [1………,M]→[1…….,N],
then a Horizontal seam is
Sy={Sjy}j=1N={y(j),j}j=1N |y(j)-y(j-1)≤j|
B.SPIHT codec
This page presents the powerful wavelet-based image
compression method called Set Partitioning in
Hierarchical Trees (SPIHT).
The SPIHT method is not a simple extension of
traditional methods for image compression, and
represents an important advance in the field. The
method deserves special attention because it provides
the following properties: Highest Image Quality,
Progressive image transmission, Fully embedded
coded file, Simple quantization algorithm, Fast
coding/decoding, Completely adaptive, Lossless
compression, Exact bit rate coding, Error protection.
Encoding part
Original input
image
Seam carving
Decoding part
Displaying
retarged image
Adjust to the
required
resolution
SPIHT encoding
Getting the
required
resolution
Compressed
encoded
SPIHT Decoding
Fig1.Program flow of the proposed image codec
http://www.ijettjournal.org
Page 2271
International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013
^
CLB=
III.PROPOSED CODEC DESCRIPTION
In this paper, SC is applied to improve the spatial
scalability of the conventional wavelet-based SPIHT
scheme. The block diagram of the proposed codec is
shown in fig 1.which involve two major aspects: the
seam carve and SPIHT coding. In the seam carve
process block based seam search process and seam
energy map are included.
From fig 1.we can see that, proposed image codec
including encoding and decoding part. At the encoder
side original input image is considered as a whole
and not divided into two components, while SC is
performed and the resultant seam energy map is used
to guide the scanning and encoding order of the DWT
coefficients. The SPIHT coded bit stream and the side
information of the resultant seams are transmitted to
the decoder side, In this way, we can reconstruct the
content aware image with arbitrary aspect ratio.
A. Compression and enlargement oriented seam
energy map
In original SC,a vertical or horizontal seam is defined
an eight connected path of pixels ,containing only one
pixel in each row or column. Good retargeting
performance can be achieved by removing or inserting
epicture compared to the original image the seam
carve added pixel to the original image and finally we
getting the retargeted image. If the user required low
resolution picture the seam carve remove pixel from
the original image and adjust to the required resolution
finally obtain the retargeted image. a forwardcumulative cost matrix can be used to search the
block-based seam path. For the vertical seams, each
entry in
is updated by
+min{MB((i-2L)/2L,(j-2L)/2L)+CLB
MB((i-2L)/2L,(j)/2L)+CUB
{ MB((i-2L)/2L,(j+2L)/2L)+CRB
Where CLB, CUB, CRB are the costs for the three
possible connection paths of a block-based vertical
seam.
CLB=
^
^
|I(i − 1, k) − I(i, k + 2^L)|
In W (.) is a weighting parameter that can be used to
balance seam removal/insertion and the important
semantic information preservation. Practically, for an
input image, visual attention analysis is first
performed using saliency detection algorithms with
manual tuning, and the ROIs and non- ROIs can be
extracted. For each ROI, the size of the bounding box
is calculated and then a corresponding key region can
be defined to cover the ROI and part of the
background. A relatively higher value of the
weighting factor can be assigned for the key regions,
while the lower is set for the rest of the image. By
using and the block-based vertical seam with
minimized forward energy can be calculated using
dynamic programming (and the calculation of the
block-based horizontal seam is similar). It is obvious
that a different size of the block in each vertical (or
horizontal) seam can generate different amount of
forward energy, since the inserted forward energy is
due to new edges induced by the previously
nonadjacent pixels that become neighbors once a seam
is removed. Generally speaking, a larger block size of
a seam unit results in higher forward energy and,
consequently, worse retargeted image quality.
However, on the other hand, as the size of the block
increases, the transmitted overhead of the position
information of the corresponding seam paths reduces.
Here in after, we optimize the block size of the blockbased seam by fully considering the retargeted image
quality and the required bit rate for coding the side
information of the seam energy map.
A cost function with a Lagrange multiplier is utilized,
and it is formulized as follows:
SBe=arg min{EB(SB)+λRB(SB)}
MB (i/2L,j/2L)=W(i/2L,j/2L)
{
∑
|I(K, j + 2 ) − I(K, j − 1)| +
Where
= 2 × 2 denotes the width (and height)
in seam unit of the block-based vertical seam, is the
optimized block size, and
are the average
and
forward seam energy induced by removing a block
based vertical seam and the required bit stream to
transmit the pixel value (or transformed coefficients)
and side information of the seams. For an N M
original image, and
can be calculated as
L
EB(SB)=∑JMB(N/2 ,j)/(M/2L)
|I(K, j + 2 ) − I(K, j − 1)| +
RB(SB)= Rc(SB)+ Rs(SB)
CLB=
^
^
|I(i − 1, K + 2 ) − I(i, k)|
|I(K, j + 2 ) − I(K, j − 1)|
ISSN: 2231-5381
Where MB(N/2 L,J) represents the last row of the cost
matrix
in, J is the column index and J
[0,M/2 ];
(
) and
http://www.ijettjournal.org
(
) are the coded bits
Page 2272
International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013
generated by encoding the wavelet coefficients and
the position information of the corresponding seam
paths, respectively. The details of
( ) and
( ) are to be introduced in the next two subsections.
Note that in ≥ 0 is the Lagrangian parameter,
and, based on the derivation in the optimal can be
set as
=-d[EB(SB)/d[RB(SB)]
In d [.] notes the derivative operation and, in practice,
the negative slope of the R-D function can be
calculated in various test images and the average value
of the slopes is employed as λ Combining the optimal
block-based seams can be found, and, in pseudo code
form, this process is illustrated . It should be
mentioned that the block-based seams are the
optimized tradeoff between the retargeting
performance and the coding efficiency. After seam
search, a block-based seam energy map can be
generated by sorting the resultant seams in energy
descending order and this map would be used to
control the scanning and coding order of the wavelet
coefficients (sorted as SOTs). B.Content Based SPIHT
Coding
Fig3: Reconstructed image with 300x300 resolution
IV.SIMULATIONS AND ANALYSIS
Here, we report the experiments conducted to evaluate
the spatial scalability, compression and enlargement
performance of the proposed seam-SPIHT.the original
input image is transferred to the seamcarving.the
seamcarving adjust the pixel to 512×512.and the
image is rotate both 90 degree and 270 degree. the
resize image is transferred to the SPIHT encoding and
compress the encoded image and the compressed
encoded image is transferred to the SPIHT decoding
part ,and getting the required resolution and adjust to
the required resolution after adjusting displaying and
saving the image. using seam-SPIHT the image can be
reconstructed with arbitrary resolution
Fig4: Reconstructed image with 2000x2000resolution
Fig5: Reconstructed image with 500x900 resolutions
Fig2:Orginal input image with 600x800 resolution
ISSN: 2231-5381
http://www.ijettjournal.org
Page 2273
International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013
CONCLUSION
This paper has presented a brief overview of
content-based image retrieval area for compression
and enlargement. we have provided a resolution for
delivering images to the diversity and versatility of
display devices with arbitrary resolutions, and
mentaning the important and sensitive content
without extra computational burden at the receiving
end. and keeps the decoder’s complexity low. In this
paper we are using seam carve and SPIHT coded
techanique.In seam-SPIHT, at the encoder side block
based seam energy map is generated in the image
domain. According to the resultant map, the
coefficients are grouped as seam-guided SOTs, Which
are encoded in energy descendant order and the side
information is also sent to indicate the positions of
trees. In this way, one can reconstruct the arbitrary
size image in a content-aware manner. Experimental
results have shown that the retargeted images
generated by the proposed codec preserve important
and sensitive image content (i.e., in a content-aware
manner), while achieving better compression and
enlargement performance compared with the existing
relevant coding schemes.
REFERENCES
[1] A. Said and W. A. Pearlman, “A new, fast and
efficient image codec based on set partitioning in
hierarchical trees,” IEEE Trans. Circuits Syst. Video
Technol., vol. 6, no. 3, pp. 243–250, Jun. 1996.
[2] C. Christopoulos, J. Askelof, and M. Larsson,
“Efficient methods for encoding regions of interest in
the upcoming JPEG 2000 still image coding
standard,” IEEE Signal Process. Lett., vol. 7, no. 9,pp.
247–249, Sep. 2000.
[3] D. Liu, X. Y. Sun, F.Wu, S. P. Li, and Y. Q.
Zhang, “Image compression with edge-based
inpainting,” IEEE Trans. Image Process., vol. 17,
[4] D. T. Vo, J. Sole, P. Yin, C. Gomila, and T. Q.
Nguyen, “Selective data pruning-based compression
using high-order edge-directed interpolation,” IEEE
Trans. Image Process., vol. 19, no. 2, pp. 399–409,
Feb. 2010. no. 10, pp. 1273–1287, Oct. 2007.
[5] S. Avidan and A. Shamir, “A seam carving for
content-aware image resizing,” ACM Trans. Graphics,
vol. 26, no. 3, pp. 10–19, Jul. 2007.
[6] N. T. N. Anh, W. X. Yang, and J. F. Cai, “Seam
carving extension: A compression perspective,” in
Proc. ACM Conf. Multimedia, Oct. 2009, pp. 825–
828.
[7] Y. Tanaka, M. Hasegawa, and S. Kato, “Image
coding using concentration and dilution based on
seam carving with hierarchical search,” in Proc. IEEE
Int. Conf. Acoust., Speech. Signal Process., Mar.
2010, pp. 1322–1325.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 2274