B070306010

International Journal of Engineering Research and Development
e-ISSN: 2278-067X, p-ISSN: 2278-800X, www.ijerd.com
Volume 7, Issue 3 (May 2013), PP. 06-10
6
Image Compression Algorithms under JPEG with Lapped
Orthogonal Transform and Discrete Cosine
Transformation
Anurag Bhatt 1
, Dr. Ashutosh Kumar Bhatt 2
1
M.Tech CSE Department, Uttarkhand Technical University, Dehradun (Uttarakhand), India
2Associate Professor.CSE Dept, Birla Institute of Applied Sciences (BIAS), Bhimtal Uttarakhand - India
Abstract:- This Paper Focus On Image Compression.It is used specially for the compression of images
where tolerable degradation is required. With the wide use of computers and consequently need for
large scale storage and transmission of data, efficient ways of storing of data have become necessary.
Image compression is minimizing the size in bytes of a graphics file without degrading the quality of
the image to an unacceptable level. It also reduces the time required for images to be sent over the
Internet or downloaded from Web pages.JPEG and JPEG 2000 are two important techniques used for
image compression.JPEG image compression standard use dct (discrete cosine transform). Now here
Lapped Orthogonal Transform and Discrete Cosine Transformation are using with JPEG 2000
standard. The discrete cosine transform is a fast transform. It is a widely used and robust method for
image compression. It has excellent compaction for highly correlated data.DCT has fixed basis images
DCT gives good compromise between information packing ability and computational complexity
Keywords:- JPEG,JPEG2000,DCT
I. INTRODUCTION
An image may be defined as a two-dimensional function f(x, y), where x and y are spatial (plane)
coordinates, and the amplitude of f at any pair of coordinates (x, y) is called the intensity or gray level of the
image at that point [8]. When x, y, and the amplitude values of f are all finite, discrete quantities, we call the
image a digital image. The field of digital image processing refers to processing digital images by means of a
digital computer. Note that a digital image is composed of a finite number of elements, each of which has a
particular location and value. These elements are referred to as picture elements, image elements, pels, and
pixels. Pixel is the term most widely used to denote the elements of a digital image.
Fig1: General compression model
Fig2: a) Source encoder model b) Source decoder model

Image Compression Algorithms under JPEG with Lapped Orthogonal Transform...
7
II. CLASSIFICATION OF IMAGE COMPRESSION SCHEMES
Image compression using transform coding yields extremely good compression, with controllable degradation of
image quality. These are:
 Lossless technique
Lossy technique
A. Lossless compression technique:
In lossless compression techniques, the original image can be perfectly recovered from the compressed
(encoded) image. These are also called noiseless since they do not add noise to the signal (image).It is also
known as entropy coding since it use statistics/decomposition techniques to eliminate/minimize Redundancy.
Lossless compression is used only for a few applications with stringent requirements such as medical imaging .
Following techniques are included in lossless compression:
1. Run length encoding 2. Huffman encoding 3. LZW coding 4. Area coding
B .Lossy compression technique:
Lossy schemes provide much higher compression ratios than lossless schemes. Lossy schemes are
widely used since the quality of the reconstructed images is adequate for most applications .By this scheme, the
decompressed image is not identical to the original image, but reasonably close to it.In this prediction –
transformation – decomposition process is completely reversible .The quantization process results in loss of
information. The entropy coding after the quantization step, however, is lossless. The decoding is a reverse
process. Firstly, entropy decoding is applied to compressed data to get the quantized data. Secondly, de
quantization is applied to it & finally the inverse transformation to get the reconstructed image . Major
performance considerations of a lossy compression scheme include:
1. Compression ratio 2. Signal - to – noise ratio 3. Speed of encoding & decoding.
Lossy compression techniques includes following schemes:
1. Transformation coding 2. Vector quantization 3. Fractal coding 4. Block Truncation Coding
5. Subband coding
III. LAPPED ORTHOGONAL TRANSFORMATION
The Lapped Orthogonal Transform denotes a lapped transform whose input length is equal to twice its
output length (L = 2M). Previous research showed the LOT is more appropriate for image coding, comparing
with other lapped transforms, because the LOT achieves higher compression efficiency for images and has less
computational cost8. 2.2.2. Design of the LOT In order to make the design of LOT robust, we can begin with a
family of valid LOTs. Then, the optimal LOT within this family or subspace can be found. Our first valid LOT
matrix3 is P0Input image is first portioned into 8x8 non overlapping blocks, and then 16x16 overlapping blocks
are obtained as shown in fig 2, since, direct scomputation of 2D LOT and its inverse requires large computation
time, a computationally efficient method described below, is used.
F=PtXP=ZtP
Where Z=XtP and Zt is its transpose. X and F are 16X16 input pixel and 8x8 coefficient matrix
IV. DISCRETE COSINE TRANSFORM
A discrete cosine transform (DCT) helps separate the image into parts (or spectral sub-bands) of
differing importance (with respect to the image's visual quality). The DCT is similar to the discrete Fourier
transform: it transforms a signal or image from the spatial domain to the frequency domain[8][9].
The basic operation of the DCT is as follows:
 The input image is N by M;
 f(i,j) is the intensity of the pixel in row i and column j;
 F(u,v) is the DCT coefficient in row k1 and column k2 of the DCT matrix.
 For most images, much of the signal energy lies at low frequencies; these appear in the upper left corner of
the DCT.
 Compression is achieved since the lower right values represent higher frequencies, and are often small -
small enough to be neglected with little visible distortion.
 The DCT input is an 8 by 8 array of integers. This array contains each pixel's gray scale level;
 8 bit pixels have levels from 0 to 255.

8
V. PURPOSE ALGORITHM
Step 1: Read Input image
Step 2: image to columns
Rearrange image blocks into columns. Image size of 512X512 now arranges in single column
i.e.272144 and transformation is applied.
B = im2col(A,[m n],block_type)
Rrearranges image blocks into columns. block_type is a string that can have one of these values. The default
value is enclosed in braces ({}
>> whos X
Name Size Bytes Class Attributes
X 1x262144 2097152 double
Step 3: Apply the Quantization and inverse Quantization technique using the threshold values.
Quantization leads to a compression of the image. Indeed, with a fixed length binary code, 8 bits per pixel are
needed to code 256 colors and 4 bits per pixel to code 16 colors. We notice that the image obtained after
quantization is of good quality. However, within the framework of true compression, quantization is not used on
the original image, but on its wavelet decomposition.
Step4 : Apply the inverse Quantization technique without any threshold values
Step 5: Rearrange matrix columns into blocks
A = col2im(B,[m n],[mm nn],'distinct') rearranges each column of B into a distinct m-by-n block to create the
matrix A of size mm-by-nn. If B = [A11(:) A21(:) A12(:) A22(:)], where each column has length m*n, then A =
[A11 A12; A21 A22] where each Aij is m-by-n.
A = col2im(B,[m n],[mm nn],'sliding') rearranges the row vector B into a matrix of size (mm-m+1)-by-(nn-n+1).
B must be a vector of size 1-by-(mm-m+1)*(nn-n+1). B is usually the result of processing the output of
im2col(...,'sliding') using a column compression function (such as sum).
col2im(B,[m n],[mm nn]) is the same as col2im(B, [m n], [mm nn],'sliding').
Step 6: the Lapped Orthogonal Transform and Discrete Cosine Transformation
Image divided into 3 sections and apply the Lapped Orthogonal Transform and Discrete Cosine Transformation
simlateneusly.
Step 7: restructure between matrix and unit range of sequences
Find the Synthesis matrix for the lapped orthogonal and Discrete Cosine Transformation and then use the
concept of reshape.
B= reshape(A ,m, n) returns the m-by-n matrix B whose elements are taken column-wise from A.
Step 8: Estimate bits for coding a cell-array of integer sequences
this estimate bits based on zero-order entropy of each sequence and entropy coding method is Huffo6.
E = entropy(I) returns E, a scalar value representing the entropy of grayscale image I. Entropy is a statistical
measure of randomness that can be used to characterize the texture of the input image. Entropy is defined as
-sum(p.*log2(p))
where p contains the histogram counts returned from imhist. By default, entropy uses two bins for logical arrays
and 256 bins for uint8, uint16, or double arrays.

9
I can be a multidimensional image. If I has more than two dimensions, the entropy function treats it as a
multidimensional grayscale image and not as an RGB image.
Step9: find out the bits and calculate the total bytes
a = numel(find(counts));
e = entropy0(counts);
b = ceil(numel(xC{i})*e) + ceil(a*log2(a));
Step 10 : Find out bit rate on the basic of transformation the
Bit rate = [0.25, 0.5, 0.75, 1.0, 1.25, 1.50, 1.75, 2.0];
Step 11: Calculation of PSNR:
R=double(input image+128)-compressed image;
PSNR = 10*log10(((numel(A)*peak^2)/sum(sum(R.*R))));
VI RESULTS
A. Peak Signal to Noise Ratio (PSNR) and Comparison
The results of table 1 have been compared with conclusion from here is the value of PSNR is good of our
proposed method for bit rates 0.6 (bpp).
Table 1: comparison study
Transformation PSNR
DCT 38.49
LOT 38.34
Hybrid(JPG) 36.76
The proposed coding scheme has been simulated on a pc with matlab. The original image 512x512,
256 gray level test image utu.bmp and size is 257 kb and after apply the hybrid technique after compressed
image size is 64.4 kb and dimension is 512x512 with PSNR 36.76 at bit rates (bits per pixel, bpp) 0.6. The
PSNR(db) of the reconstructed images were recorded as a function of bit rate in bits per pixel (bpp), which is
determined from the actual size in bytes from the compressed files.
The principles and performance of the Lapped Orthogonal Transform and Discrete Cosine
Transformation are discussed here. Lapped Orthogonal Transform and Discrete Cosine Transformation delivers
higher quality images at various bit rates than the traditional DCT transform coding schemes. Extremely low bit
rate as 0.10 bpp or 0.15 bpp has little importance for implementations. Compressing images in the range of rates
from 0.25 to 0.75 bpp seems to be the ideal place to enjoy the advantage of the Lapped Orthogonal Transform
and Discrete Cosine Transformation On the other hand, this scheme only increases computational cost by a
small amount over that of the DCT

10
VII CONCLUSION
In this Paper image compression techniques using DCT and LOT were implemented. DCT is used for
transformation in JPEG standard.DCT performs efficiently at medium bit rates. 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 decor related at their boundaries using DCT. DCT is
used as basis for transformation in JPEG 2000 standard. DCT provides high quality compression at low bit rates.
Though the DCT-based image compression method performs well at moderate bit a rate, the image
quality degrades rapidly at higher compression ratios. This is due to the artifacts resulting from the block-based
DCT scheme. On the contrary, dct and lot based compression technique provides substantial enhancement in
picture quality at low bit rates due to overlapping basis functions and better energy compaction property of
wavelet transforms. This thesis presents some mathematical background of compression technique mainly based
on the transformation as well as the compression method.
We have introduced a block-based image compression algorithm capable of exploiting the transformation
technique. For this purpose, we presented a transform design method that jointly optimizes the classification of
blocks and corresponding transforms lot and dct. The result is a set of optimal transforms that replace the
traditional block transforms used in image compression. The geometry information is used only at the
initialization of the transform optimizations. Our compression results show PSNR against thee bits rate based on
entropy method Huffo6.
REFERENCES
[1]. Ahmed, N., Natarajan, T., Rao, K. R., “Discrete Cosine Transform”, IEEE Trans. Computers, vol. C-23,
Jan. 1974, pp. 90-93.
[2]. Watson, A. B. DCT quantization matrices visually optimized for individual images. Proceedings of the
SPIE 1913: 202-216 (Human Vision, Visual Processing, and Digital Display IV. Rogowitz ed. SPIE.
Bellingham, WA). 1993
[3]. Matthias Kramm, "Compression of Image Clusters using Karhunen Loeve Transformations", TU-
M¨unchen, Institute for Computer Science, Boltzmannstr. 3, 85748 Garching, Germany.
[4]. Chengjie Tu, And Trac D. Tran, "Context-Based Entropy Coding of Block Transform" Coefficients
for Image Compression, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 11, NO. 11,
NOVEMBER 2002, PP-1271-1283
[5]. Sugreev Kaur and Rajesh Mehra, “HIGH SPEED AND AREA EFFICIENT 2D DWT PROCESSOR
BASED IMAGE COMPRESSION Signal & Image Processing : An International Journal(SIPIJ) Vol.1,
No.2, December 2010.
[6]. Zhigang Gao, Yuan F. Zheng, “ Quality Constrained Compression Using DWT Based Image Quality
Metric”, Dept. of Electrical and Computer Engineering The Ohio State University
[7]. Kamrul Hasan Talukder and Koichi Harada, “Discrete Wavelet Transform for Image Compression and
A Model of Parallel Image Compression Scheme for Formal Verification”, , Proceedings of the World
Congress on Engineering 2007 Vol I WCE 2007, July 2 - 4, 2007, London, U.K.
[8]. Liu Chien-Chih, Hang Hsueh-Ming, "Acceleration and Implementation of JPEG 2000 Encoder on TI
DSP platform" Image Processing, 2007. ICIP 2007. IEEE International Conference on, Vo1. 3, pp. III-
329-339, 2005
[9]. J. Padhye, V. Firoiu, and D. Towsley, “A stochastic model of TCP Reno congestion avoidance and
control,” Univ. of Massachusetts, Amherst, MA, CMPSCI Tech. Rep. 99-02, 1999.Wireless LAN
Medium Access Control (MAC) and Physical Layer (PHY) Specification, IEEE Std. 802.11, 1997.
[10]. H. S. Malvar and D. H. Staelin. “The LOT: Transform coding without blocking effects,” IEEE Trans.
Acoustics, Speech, Signal Processing, vol. 37, April 1989, pp. 553 – 559.

B070306010

More Related Content

What's hot (18)

Viewers also liked (15)

Similar to B070306010 (20)

More from IJERD Editor (20)

Recently uploaded (20)

B070306010