Project

WELCOME
IMAGE COMPRESSION
INTRODUCTION
 Uncompressed multimedia data requires

considerable storage capacity and transmission
bandwidth.
 For still image compression JPEG standard has

been established. The performance of these codes
degrades at low bit rates
INTRODUCTION
 The wavelet transform has emerged within the field of

image compression.
 Wavelet based coding provides substantial

improvements in picture quality at higher
compression ratios
IMAGE COMPRESSION
Compression is basically of two types.
 Lossy Compression
 Lossless Compression.
LOSSY COMPRESSION
 Under normal viewing conditions no visible loss
 A certain loss of accuracy
 Effective when applied to graphics images and

digitized voice.
LOSSLESS COMPRESSION
 An exact duplicate of the input data stream
 The reconstructed image is numerically identical

to the original image
 Only a modest amount of compression.
 Used when storing data base records, spread

sheets or word processing files
BLOCK DIAGRAM
SOURCE ENCODER
 A linear transformer
 Compression using wavelet transforms-

transform coding
WHY TRANSFORM ??

To obtain further information of the signal
DIFFERENT TYPES OF TRANSFORMS:
Fourier Transform

Short time fourier transform


Wavelet transform

FOURIER TRANSFORM
Can determine frequency contents of signals.
No time information in frequency domain
No frequency information in time domain

SHORT TIME FOURIER TRANSFORM
Signal is divided in to small segments by multiplying


with window functions
Constant resolution at all frequencies and

times
WAVELET TRANSFORM
CONTINUOUS WAVELETTRANSFORM
All wavelet derived from mother wavelet
Inverse Wavelet Transform

Admissibility condition
DISCRETE WAVELET TRANSFORM
DWT of a signal can be calculated by passing

it through a series of filters
TIME FREQUENCY PLANE
STFT WT
CLASSIFYING IMAGE DATA
 An image is represented as a two dimensional array of

coefficients
 The smooth variations in colour can be termed as low
frequency variations and the sharp variations as high
frequency variations.
 The low frequency components constitute the base of
an image
CLASSIFYING IMAGE DATA
 The smooth variations are demanding more
importance than the details.
 Separating the smooth variations and details of the
image can be done using Discrete Wavelet Transform
(DWT).
DWT OF AN IMAGE
 A low pass filter and a high pass filter are chosen
 They can be sub sampled by two
 The output data now contains only half the original
number of samples
 High pass filter is applied for the same row of data
 The filtering is done for each column of the
intermediate data
SINGLE LEVEL DECOMPOSITION
LEVEL OF DECOMPOSITION
SINGLE LEVEL DECOMPOSITION

TWO LEVEL DECOMPOSITION
THREE LEVEL DECOMPOSITION

INVERSE DWT OF AN IMAGE
 Used to reassemble the various
classes of data into a reconstructed image
 A pair of high pass and low pass filters is used
 The filtering procedure is just the opposite
RECONSTRUCTION
QUANTIZATION
 Approximating the continuous set of values in the
image data with a finite set of values
 Output values are discrete, and usually finite.
QUANTIZER
uniform quantizer
non-uniform quantizer
DEQUANTIZER
 Receives quantizer output and converts them
into normal data
ENTROPY CODING
 To give additional compression

 Encodes the given set of symbols with the
minimum number of bits required to represent
them
ADVANTAGES
 The high image compression ratios reduces the
hard disk storage capacity
 More robust under transmission errors
 The compressed video file cannot be edited
 Because wavelet transforms compress the entire frame,
any change makes it impossible to decompress the
image
APPLICATIONS
 JPEG2000 uses wavelet transforms to compress
images
 MPEG-4 uses wavelet tiling to allow the division of
images into several tiles, each with separate
encoding
CONCLUSION
 Substantial improvement in picture quality at low bit
rates
“THANK
YOU”

Project

Uploaded by

Copyright:

Available Formats

Project

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Project

Uploaded by

Copyright:

Available Formats

WELCOME

 Uncompressed multimedia data requires

 For still image compression JPEG standard has

 The wavelet transform has emerged within the field of

 Wavelet based coding provides substantial

Compression is basically of two types.

 Under normal viewing conditions no visible loss

 A certain loss of accuracy

 Effective when applied to graphics images and

 The reconstructed image is numerically identical

 Only a modest amount of compression.

 Used when storing data base records, spread

 Compression using wavelet transforms-

DIFFERENT TYPES OF TRANSFORMS:

Short time fourier transform

Can determine frequency contents of signals.

No time information in frequency domain

No frequency information in time domain

Signal is divided in to small segments by multiplying

with window functions

Constant resolution at all frequencies and

All wavelet derived from mother wavelet

Inverse Wavelet Transform

DISCRETE WAVELET TRANSFORM

DWT of a signal can be calculated by passing

 An image is represented as a two dimensional array of

SINGLE LEVEL DECOMPOSITION

THREE LEVEL DECOMPOSITION

 To give additional compression

You might also like