Professional Documents
Culture Documents
Digital Assignment I
Digital Assignment I
Digital Assignment - I
Ans: We can achieve compression by removing of one or more of three basic data
redundancies:
Literature Survey
3) Image compression using discrete cosine transform & discrete wavelet transform by
Bhawna Gautam
Image compression is very important for efficient transmission and storage of images. Demand
for communication of multimedia data through the telecommunications network and accessing
the multimedia data through Internet is growing explosively. With the use of digital cameras,
requirements for storage, manipulation, and transfer of digital images, has grown explosively.
These image files can be very large and can occupy a lot of memory. A Gray scale image that
is 256 x 256 pixels has 65, 536 elements to store, and a typical 640 x 480 colour image has
nearly a million. Downloading of these files from internet can be very time-consuming task.
Image data comprise of a significant portion of the multimedia data and they occupy the major
portion of the communication bandwidth for multimedia communication. Therefore,
development of efficient techniques for image compression has become quite necessary. A
common characteristic of most images is that the neighbouring pixels are highly correlated and
therefore contain highly redundant information. The basic objective of image compression is
to find an image representation in which pixels are less correlated. The two fundamental
principles used in image compression are redundancy and irrelevancy. Redundancy removes
redundancy from the signal source and irrelevancy omits pixel values which are not noticeable
by human eye. JPEG and JPEG 2000 are two important techniques used for image
compression.
Methods / Result
JPEG stands for the Joint Photographic Experts Group, a standards committee that had its
origins within the International Standard Organization (ISO).JPEG provides a compression
method that is capable of compressing continuous-tone image data with a pixel depth of 6 to
24 bits with reasonable speed and efficiency.JPEG may be adjusted to produce very small,
compressed images that are of relatively poor quality in appearance but still suitable for many
applications. Conversely, JPEG is capable of producing very high-quality compressed images
that are still far smaller than the original uncompressed data.
Results obtained after performing DCT of various orders on original images are shown below:
Conclusion
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 neighbouring blocks is
neglected. Blocks cannot be decorrelated at their boundaries using DCT. DWT is used as basis
for transformation in JPEG 2000 standard. 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. However, DWT provides lower quality than JPEG at low
compression rates. DWT requires longer compression time.
The rapid growth of digital imaging applications, including desktop publishing, multimedia,
teleconferencing, and high-definition television (HDTV) has increased the need for effective
and standardized image compression techniques. Among the emerging standards are JPEG, for
compression of still images [Wallace 1991]; MPEG, for compression of motion video [Puri
1992]; and CCITT H.261 (also known as Px64), for compression of video telephony and
teleconferencing. All three of these standards employ a basic technique known as the discrete
cosine transform (DCT). Developed by Ahmed, Natarajan, and Rao [1974], the DCT is a close
relative of the discrete Fourier transform (DFT). Its application to image compression was
pioneered by Chen and Pratt [1984]. In this article, some simple functions will be developed to
compute the DCT and show how it is used for image compression. We have used these
functions in our laboratory to explore methods of optimizing image compression for the human
viewer, using information about the human visual system [Watson 1993]. The goal of this paper
is to illustrate the use of Mathematica in image processing and to provide the reader with the
basic tools for further exploration of this subject.
Methods / Result
DCT-based image compression relies on two techniques to reduce the data required to represent
the image. The first is quantization of the image's DCT coefficients; the second is entropy
coding of the quantized coefficients. Quantization is the process of reducing the number of
possible values of a quantity, thereby reducing the number of bits needed to represent it.
Entropy coding is a technique for representing the quantized data as compactly as possible. We
will develop functions to quantize images and to calculate the level of compression provided
by different degrees of quantization. We will not implement the entropy coding required to
create a compressed image file.
Conclusion
Most of the computation time required to transform, quantize, dequantize, and reconstruct an
image is spent on forward and inverse DCT calculations. Because these transforms are applied
to blocks, the time required is proportional to the size of the image. On a SUN SPARCstation
2, the timings increase (at a rate of 0.005 second/pixel) from about 20 seconds for a 642-pixel
image to about 320 seconds for 2562 pixels. These times are much longer than for comparable
functions written in a low-level language such as C. For example, a C program performed the
same computations in under 2 seconds for an image of 2562 pixels, more than 100 times faster
than our Mathematica functions. However, for the purposes for which our code was developed,
namely education, algorithm development, and prototyping other applications, the timings are
acceptable.
In recent years there has been an increasing interest with respect to using a class of orthogonal
transforms in the general area of digital signal processing. This correspondence addresses itself
towards two problems associated with image processing, namely, pattern recognition and
Wiener filtering. In pattern recognition, orthogonal transforms enable a noninvertible
transformation from the pattern space to a reduced dimensionality feature space. This allows a
classification scheme to be implemented with substantially less features, with only a small
increase in classification error. In discrete Wiener filtering applications, the filter is represented
by an (M X M) matrix G. The estimate X of data vector X is given by GZ, where Z = X + N
and N is the noise vector. This implies that approximately 2M arithmetic operations are
required to compute X. Use of orthogonal transforms yields a G in which a substantial number
of elements are relatively small in magnitude, and hence can be set equal to zero. Thus, a
significant reduction in computation load is realized at the expense of a small increase in the
mean-square estimation error.
Methods / Result
The minimization of a linear sequential machine to a reduced form is an important topic and is
discussed in many texts. The minimization procedure presented is as follows. Let {A, B, C, D}
be an n-dimensional linear machine over GF(p), and let r, with r < n, be the rank of the
diagnostic matrix K A [C'A'C' … (A’) n-1 C']’, where the prime stands for the transpose. Define
an r X n matrix T consisting of the first r linearly independent rows of K, and an n X r matrix
R denoting the right inverse of T so that TR = L Then the linear machine {TAR, TB, CR, D}
is a reduced form of {A, B, C, D}. In this correspondence, an algorithm will be introduced to
find a special set of r linearly independent rows in K. A reduced machine can then be read out
from this algorithm without the need of inverting any matrix. Furthermore, the reduced
machine will be in a canonical form.
Conclusion
A new algorithm for computing the correlation of a block of sampled data has been presented.
It is a direct method which trades an increased number of additions for a decreased number of
multiplications. For applications where the "cost" (e.g., the time) of a multiplication is greater
than that of an addition, the new algorithm is always more computationally efficient than direct
evaluation of the correlation, and it is generally more efficient than FFT methods for processing
128 or fewer data points, or for calculating only the first L "lags" for L < 10 log2 2N.
This paper presents a multimedia processor based on a SIMD architecture optimized for block-
based video processing algorithms. The processor, called DGP (Digital Generic Processor), is
a generic system architecture, constituted by an array of pixel processors (SZMD) and a RISC
controller. It is able to execute various video processing algorithms such as digital filtering,
video effects, window clipping, and to perform video compression according to standards such
as H.261, H.263, MPEG-I and MPEG-2. The processor can be programmed in order to run the
code corresponding to each specific algorithm.
Methods / Result
Three main parts constitute the architecture of the DGP:
• a SIMD architecture composed of 64-pixel processors connected by a programmable
network.
• a 32-bit core RISC controller (master processor) which performs interfaces and
transfers exchange data with the SIMD array.
• A variable length codes unit (VLC-Coprocessor) which can load, encode and decode
variable length codes or arithmetic codes.
The master processor and the SIMD engine have their own data transfer unit based on a 32-bit
wide and its associated address is 16 bits wide. All units are connected to that bus. The master
controller is the master of that system bus and he manages also the VLD-coprocessor when
incoming data are needed for some algorithms.
Conclusion
A fixed array of a small number of cameras positioned to view a work environment or
demonstration area can be used to synthesize new perspective views of that environment. The
geometry of the fixed cameras will determine the extent to which new views can be created.
Depth resolution is limited by the pixel size, the width if the images and the separation of the
camera viewpoints. For many interactive applications, it is not necessary to fully define an
explicit three-dimensional model, and implicit models in which the depth of significant points
is determined may be adequate. Decreasing costs of computational power and imaging devices
make this type of interaction feasible, and the structure of the computation is well matched to
parallel implementations.
Conclusion
From the experimental results, we can see that the compression ratio using DWT is greater.
Which means compression by using DWT method is smaller. However, we can get a better
image with DCT method.
DCT (Discrete Cosine Transform)
Concentrate image energy into a small number of coefficients (energy compaction).
Minimizes interdependencies among coefficients (decorrelation).
Not resistant to changes in an object
DCT calculates the quantity of bits of image where the message is hidden inside.
Ans: The first step is to convert an image to Y’CbCr and just pick the Y’ channel
and break into 8 x 8 blocks. Then starting from the first block, map the range from
-128 to 127. After that you have to find the discrete Fourier transform of the
matrix. The result of this should be quantized. The last step is to apply encoding
in the zig zag manner and do it till you find all zero.
Note. You have to repeat this procedure for all the block of 8 x 8.
The following is a general overview of the JPEG process.
1. The image is broken into 8x8 blocks of pixels.
2. Working from left to right, top to bottom, the DCT is applied to each
3. Each block is compressed through quantization. Block
4. The array of compressed blocks that constitute the image is stored in a
drastically reduced amount of space
5. When desired, the image is reconstructed through decompression, a process
that uses the Inverse Discrete Cosine Transform (IDCT).
References
1. Raid, A. M., Khedr, W. M., El-Dosuky, M. A., & Ahmed, W. (2014). Jpeg image compression using
discrete cosine transform-A survey. arXiv preprint arXiv:1405.6147.
2. Mathey, R., & Avadhani, P. S. An Multi Resolution Using Discrete Wavelet Transforms and Fractals
Transforms.
3. Bhawna, G. (2010). Image compression using discrete cosine transform and discrete wavelet
transform (Doctoral dissertation).
4. Watson, A. B. (1994). Image compression using the discrete cosine transform. Mathematica
journal, 4(1), 81.
5. Ahmed, N., Natarajan, T., & Rao, K. R. (1974). Discrete cosine transform. IEEE transactions on
Computers, 100(1), 90-93.
6. Boliek, M., Gormish, M. J., Schwartz, E. L., & Keith, A. (1997, October). Next generation image
compression and manipulation using CREW. In Proceedings of International Conference on Image
Processing (Vol. 2, pp. 567-570). IEEE.
7. Kurniawan, A., Purboyo, T. W., & Prasasti, A. L. (2017). Implementation of image compression
using discrete cosine transform (dct) and discrete wavelet transform (dwt). International Journal of
Applied Engineering Research, 12(23), 13951-13958.