Research Journal of Applied Sciences, Engineering and Technology 4(24): 5515-5518, 2012
ISSN: 2040-7467
© Maxwell Scientific Organization, 2012
Submitted: March 23, 2012 Accepted: April 30, 2012 Published: December 15, 2012
Corresponding Author: K. Dinesh, J. P College of Engineering, Ayikudi, Tirunelveli, Tamilnadu, India
5515
Color Image and Video Compression Based on Direction Adaptive Partitioned
Discrete Wavelet Transform
K. Dinesh, S. Saravana Kumar and P. Daniel
J.P. College of Engineering, Ayikudi, Tirunelveli, Tamilnadu, India
Abstract: The main objective of this study is to use the Direction-Adaptive Partitioned Block Transform
(DA-PBT) for compressing the color images and videos. It is same as the direction-adaptive block
transform but it also have an additional direction-adaptive block partitioning to improve energy
concentration. The selection of a directional mode determines the transform direction that provides
directional basis functions. It reduces complexity and more efficient coefficient ordering for entropy
coding. For image coding, the DA-PBT significantly outperforms the directional DCT. As a block
transform, the DA-PBT can be directly incorporated into the prediction-based video coding standards to
work with the block-based intra prediction as well as the block-based motion-compensated interframe
prediction. The performance of the DA-PBT is compared with the 2D-DCT by using the Peak-Signal-to-
Noise Ratio (PSNR) and Compression Ratio (CR). The experimental results shows that the DA-PBT
performs well than the 2D-DCT.
Keywords: Compression techniques, discrete wavelet transforms, video coding standards
INTRODUCTION
Image compression has been the key technology
for transmitting massive amount of real-time image
data via limited bandwidth channels. The data are
transferred in the form of image, graphics, audio and
video. These types of data have to be compressed
during the transmission process. Otherwise the data
need a large storage capacity and transmission
bandwidth if it is uncompressed. Huge amount of data
can’t be fit if there is low storage capacity present. To
solve this problem the data has to be compressed by
using any one of the algorithms and then it can be sent
easily. Data compression is the process of converting an
input data stream into another data stream that has
smaller size. There are two types of image
compression: lossless and lossy. With lossless
compression, the original image is recovered exactly
after decompression. Much higher compression ratios
can be obtained if some error, which is usually difficult
to perceive, is allowed between the decompressed
image and the original image. This is lossy
compression. In many cases, it is not necessary or even
desirable that there be error-free reproduction of the
original image. In such a case, the small amount of
error introduced by lossy compression may be
acceptable. Another application where lossy
compression is acceptable is in fast transmission of still
images over the Internet.
For over two decades, block transforms such as the
2-D DCT have been the key component in image
coding techniques, e.g., the JPEG standard (Information
Technology, 1992). Although more recently,
convolutional transforms, such as the 2-D DWT, have
been proved superior (Taubman and Marcellin, 2000;
Information Technology, 2002), JPEG is still the
prevalent image coding format to date. Block
transforms also remain an integral part in most video
coding standards since they can be effectively
combined with block-based motion-compensated
prediction (Wiegand et al., 2003). In image coding,
directional features such as lines and edges have
significant impact on both the objective and the
subjective performance. Block transforms are typically
constructed in a separable manner, by cascading a 1-D
vertical transform and a 1-D horizontal transform.
Separable transforms tend to do well for image detail
oriented strictly horizontally and vertically, while
ringing and checkerboard artifacts are likely to appear
around edges of other orientations, significantly
deteriorating visual quality. To exploit directionality,
the intra coding portion of the video coding standard
H.264 predicts a block from previously encoded blocks
using directional extrapolation (Wiegand et al., 2003)
and the residual is then encoded via a block transform.
This approach is effective, but it requires perfect
synchronization of the DPCM loops of encoder and
decoder. Any deviation between encoder and decoder
can propagate over the image, leading to an