Chapter 8

LOSS COMPRESSION
ALGORITHM
Contents
8.1. Introduction
8.2. Distortion Measures
8.3. The Rate Distortion Theory
8.4. Quantization
8.5. Transform Coding

1/11/2023 CH-8 2
8.1. Introduction
• As discussed in Chapter 7, 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 them approximation is to the
original data, some form of distortion measure is required.

1/11/2023 CH-8 3
8.2. Distortion Measures
• A distortion measure is a mathematical quantity that specifies how close an approximation is to its
origina1, 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 were 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, not a more naive
numerical approach. Of the many numerical distortion measures that have been defined, we present the
three most commonly used in image compression. If we are interested in the average pixel difference, the
mean square error (MSE) α 2 is often used. It is defined as:

Where Xn, Yn, and N are the input data sequence,

• Where σx2 is the average square value of the original data sequence and σd2 is the
MSE. Another commonly used measure for distortion is the peak-signal-to-noise
ratio (PSNR), which measures the size of the error relative to the peak value of the
signal x peak. It is given by:

1/11/2023 CH-8 5
8.3. The Rate Distortion Theory
• Lossy compression always involves a tradeoff between rate and distortion. Rate is the
average number of bits required to represent each source symbol. Within this framework, the
tradeoff between rate and distortion is represented in the form of a rate-distortion function
R(D).
• Intuitively, for a given source and a given distortion measure, if D is a tolerable amount of
distortion, R(D) specifies the lowest rate at which the source data can be encoded while
keeping the distortion bounded above by D. It is easy to see that when D=0, we have a
lossless compression of the source. The rate-distortion function is meant to describe a
fundamental limit for the performance of a coding algorithm and so can be used to evaluate
the performance of different algorithms.
• Figure 8.1 shows a typical rate-distortion function. Notice that the minimum possible rate at
D = 0, no loss, is the entropy of the source data. The distortion corresponding to a rate R(D)
= 0 is the maximum amount of distortion incurred when “nothing” is coded. Finding a
closed-form analytic description of the rate-distortion function for a given source is difficult,
if not impossible.
1/11/2023 CH-8 6
8.3. The Rate Distortion Theory…
Figure 8. 1 Typical rate-distortion function.

1/11/2023 CH-8 7
8.4. 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 small set, via
quantization.

• Each algorithm (each quantizer) can be uniquely determined by its partition of the input range on
the encoder side and the set of output values, on the decoder side. The input and output of each
quantizer can be either scalar values or vector values, thus leading to scalar quantizer and vector
quantizer.

1/11/2023 CH-8 8
8.5. Transform Coding
• From basic principles of information theory, we know that coding vectors is more efficient than
coding scalars. To carry out such an intention, we need to group blocks of consecutive samples
from the source input into vectors.

• Let X = {X1, X2, … , Xk}T be a vector of samples. Whether our input data is an image, a piece
of music, an audio or video clip, or even a piece of text, there is a good chance that a substantial
amount of correlation is inherent among neighboring samples Xi. The rationale behind transform
coding is that if Y is the result of a linear transform T of the input vector X in such a way that the
components of Y are much less correlated, then Y can be coded more efficiently than X.

1/11/2023 CH-8 9
8.5. Transform Coding…
• For example, if most information in an RGB image is contained in a main axis, rotating so that
this direction is the first component means that luminance can be compressed differently from
color information. This will approximate the luminance channel in the eye. In higher dimensions
than three, if most information is accurately described by the first few components of a
transformed vector, the remaining components can be coarsely quantized, or even set to zero,
with little signal distortion. The more decorrelated – that is, the less effect one dimension has on
another (the more orthogonal the axes), the more chance we have of dealing differently with the
axes that store relatively minor amounts of information ,without affecting reasonably accurate
reconstruction of the signal from its quantized or truncated transform coefficients.
• Generally, the transform T itself does not compress any data. The compression comes from the
processing and quantization of the components of Y. Discrete Cosine Transform (DCT) as a tool
to decorrelate the input signal.

1/11/2023 CH-8 10
8.5. Transform Coding…
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. We will examine the definition
of the DCT and discuss some of its properties, in particular the relationship between
it and the more familiar Discrete Fourier Transform (DFT).
• Definition of DCT: Let’s start with the two-dimensional DCT. Given a function f (i,
j) over two integer variables i and j (a piece of an image), the 2D DCT transforms it
into a new function F(u, v), with integer u and v running over the same range as i and
j.

1/11/2023 CH-8 11
Review Questions
1. What is distortion?
2. What does it mean D=0?
3. How to Quantize the images? Explain it
4. By which mechanism to transform encoding is done? Explain it.

1/11/2023 CH-8 12