Scribd
Upload
9 views13 pages

Chapter 8

Chapter 8 of the Multimedia Systems course discusses loss compression algorithms, focusing on distortion measures, rate distortion theory, quantization, and transform coding. It highlights the importance of perceptual distortion in lossy compression and introduces techniques such as uniform and nonuniform scalar quantization, vector quantization, and discrete cosine transform (DCT). The chapter concludes with the Karhunen-Loeve Transform (KLT) as a method for optimal decorrelation of input data.

Uploaded by

abdlkaf.aman
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
9 views13 pages

Chapter 8

Chapter 8 of the Multimedia Systems course discusses loss compression algorithms, focusing on distortion measures, rate distortion theory, quantization, and transform coding. It highlights the importance of perceptual distortion in lossy compression and introduces techniques such as uniform and nonuniform scalar quantization, vector quantization, and discrete cosine transform (DCT). The chapter concludes with the Karhunen-Loeve Transform (KLT) as a method for optimal decorrelation of input data.

Uploaded by

abdlkaf.aman
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 13

Course Title

Multimedia Systems
CoSc4072

Chapter 8
Loss Compression Algorithms

By: Ins. Solomon S.


Outlines
 Distortion Measures
The Rate Distortion Theory
Quantization
Transform Coding
Introduction
 The compression ratio for image data using lossless compression techniques (e.g.,
Huffman Coding, Arithmetic Coding, LZW) is low when the image histogram is
relatively flat.
 For image compression in multimedia applications, where a higher compression ratio is
required, lossy methods are usually adopted.
 In lossy compression, the compressed image is usually not the same as the original
image but is meant to form a close approximation to the original image perceptually.
 To quantitatively describe how close the approximation is to the original data, some
form of distortion measure is required.
Distortion Measures
 A distortion measure is a mathematical quantity that specifies how close an approximation is to its
original, using some distortion criteria.
 When looking at compressed data, it is natural to think of the distortion in terms of the numerical
difference between the original data and the reconstructed data.
 However, when the data to be compressed is an image, such a measure may not yield the intended
result.
 For example, if the reconstructed image is the same as original image except that it is shifted to the right
by one vertical scan line, an average human observer would have a hard time distinguishing it from the
original and would therefore conclude that the distortion is small.
 However, when the calculation is carried out numerically, we find a large distortion, because of the large
changes in individual pixels of the reconstructed image.
 The problem is that we need a measure of perceptual distortion.
The rate distortion theory
 Rate–distortion theory is a major branch of information theory which provides the theoretical
foundations for lossy data compression;
 It addresses the problem of determining the minimal number of bits per symbol, as measured by the rate
R, that should be communicated over a channel, so that the source (input signal) can be approximately
reconstructed at the receiver (output signal) without exceeding an expected distortion D.
 The tradeoff between rate and distortion is represented in the form of a rate-distortion function R(D).

 It is easy to see that when D = 0, we have a lossless compression of the source.


Quantization
 Quantization in some form is the heart of any lossy scheme. Without
quantization, we would indeed be losing little information.
 The source we are interested in compressing may contain a large number
of distinct output values (or even infinite, if analog).
 To efficiently represent the source output, we have to reduce the number
of distinct values to a much smaller set, via quantization.
Uniform Scalar Quantization

A uniform scalar quantizer partitions the domain of input values into


equally spaced intervals, except possibly at the two outer intervals.
The endpoints of partition intervals are called the quantizer's decision
boundaries.
Nonuniform Scalar Quantization
 If the input source is not uniformly distributed, a uniform quantizer may
be inefficient.
 Increasing the number of decision levels within the region where the source is
densely distributed can effectively lower granular distortion.
 In addition, without having to increase the total number of decision levels,
we can enlarge the region in which the source is sparsely distributed.
 Such nonuniform quantizers thus have nonuniformly defined decision
boundaries.
Vector Quantization
 One of the fundamental ideas in Sharmon's original work on inflation theory
is that any compression system performs better if it operates on vectors or
groups of samples rather than on individual symbols or samples.
 We can form vectors of input samples by concatenating a number of
consecutive samples into a single vector.
 For example, an input vector might be a segment of a speech sample, a
group of consecutive pixels in an image, or a chunk of data in any other
format.
Transform Coding

 From basic principles of information theory, we know that coding vectors


is more efficient than coding scalars.
 To implement such intention, we need to group blocks of consecutive
samples from the source input into vectors.
Discrete Cosine Transform (DCT)

 The Discrete Cosine Transform (DCT), a widely used transform coding


technique, is able to perform decorrelation of the input signal in a data-
independent manner.
 Because of this, it has gained tremendous popularity.
Karhunen-Loeve Transform

 The Karhunen-Loeve Transfonn (KLT) is a reversible linear transform that


exploits the statistical properties of the vector representation.
 Its primary property is that it optimally decorrelates the input.
 To do so, it fits an n-dimensional ellipsoid around the (mean-subtracted) data.
 The main ellipsoid axis is the major direction of change in the data
End of Chapter 8

You might also like