23-05-2012, 11:46 AM
Compressed Image Processing
[attachment=22822]
Abstract:
Image processing has gone through many years of research and development. Many techniques/
methods have been developed which can be applied to compress images. Compressed Image
processing addresses the problem of reducing the amount of data required to represent an image.
This paper gives the overview of the invention of image compression. Also various methods of
image compression, coding the image data into a compact form are discussed.
An imaging apparatus receives a stream of image data compressed according to a token
based compression algorithm. The token dictionary is decompressed and processed according to
desired standards such as optimizing for a particular image output device. After revision, the
revised tokens are stored in a revised dictionary which is accessed upon decompression of image
content. When the image content is decompressed the locations in the compressed or coded input
data stream refer to locations in the revised dictionary which are extracted and output either to hard
copy or for further processing.
This paper reports some pioneering work in this direction to show that compressed image
processing could well become a new research area and challenge right through the next century
Image Compression
Introduction:
Image compression is minimizing the size in bytes of a graphics file without
degrading the quality of the image to an unacceptable level. The reduction in file size allows more
images to be stored in a given amount of disk or memory space. It also reduces the time required
for images to be sent over the Internet or downloaded from Web pages
Image compression is important for web designers who want to create faster loading web
pages which in turn will make your website more accessible to others. This image compression will
also save you a lot of unnecessary bandwidth by providing high-quality image with fraction of file
size.
Image compression is also important for people who attach photos to emails which will
send the email more quickly, save bandwidth costs and not make the recipient of the email angry.
Sending large image attachments can be considered offensive. This makes people very upset
because the email takes a long time to download and it uses up their precious bandwidth.
Furthermore image compression plays a major role in many important and diverse
applications including televideo conferencing, remote sensing, document and medical imaging,
hazardous waste management application and the like.
LOSSLESS AND LOSSY IMAGE COMPRESSION:
There are several different techniques in which image files can be compressed. A
text file or program can be compressed without the introduction of errors, but only up to a certain
extent. This is called lossless compression. On the other hand, lossy compression reduces a file by
permanently eliminating certain information, especially redundant information.
A typical lossy image compression system is shown in the following figure. There
are three components for the encoder namely Source Encoder, quantizer and entropy encoder.
Source encoder (linear transforms):- A variety of linear transforms have been
developed which include Discrete Fourier Transforms (DFT), discrete cosine transforms (DCT),
discrete wavelet transform (DWT) and many more each with its own advantages and
disadvantages.
Quantizer: - A quantizer simply reduces the number of bits needed to store the
transformed coefficients by reducing the precision of those values. Since this is a many-to-one
mapping, it’s a lossy process and is the main source of compression in an encoder. The
quantization performed on individual coefficient is known as scalar Quantization (SQ), and that
performed on group of coefficients together is called Vector Quantization (VQ).
Entropy encoder: - Entropy encoder further compresses the quantized values
losslessly to give better overall compression. It uses a model to accurately determine the
probabilities for each quantized value and produces an appropriate code based on these
probabilities so that the resultant output code stream will be smaller than the input stream.
PREDICTIVE AND TRANSFORM IMAGE COMPRESSION:
In predictive coding, information already sent are available is used to predict future
values, and the difference is coded. Differential Pulse Code Modulation (DPCM) is an example for
predictive coding.
In transform coding, first the image is transformed from its spatial domain
representation to a different type of representation using some well known transform and then
codes the transformed values (coefficients).
The above figure represents a typical data compression system.
Methods of compressing images:-
RUN LENGTH ENCODING (RLE)
The repeating characters are called runs. Run-Length encoding(RLE) is a very
simple form of data compression in which runs of data are stored as a single data value and count,
rather than as the original run and taking advantage of repetitive data. This is most useful on data
that contains many such runs; for example, some graphic images such as icons and line drawings.
RLE performs lossless data compression and is well suited to palette-based iconic
images. It does not work well at all on continuous-tone images such as photographs. This works
very well for images with solid backgrounds like cartoons
Properties of Image Compression:
There are a few important properties of image compression schemes:
Scalability generally refers to a quality reduction achieved by manipulation of the
bitstream or file (without decompression and re-compression). Other names for scalability are
progressive coding or embedded bitstreams. There are several types of scalability:
Quality progressive or layer progressive: The bitstream successively refines the
reconstructed image.
Resolution progressive: First encode a lower image resolution; then encode the
difference to higher resolutions.
Component progressive: First encode grey; then color.
Region of Interest Coding: Certain parts of the image are encoded with higher
quality than others. This can be combined with scalability (encode these parts first, others later).
Meta Information: Compressed data can contain information about the image which can be
used to categorize, search or browse images. Such information can include color and texture
statistics, small preview images and author/copyright information.
Processing Power: Compression algorithms require different amounts of processing power
to encode and decode. Some high compression algorithms require high processing power.
TYPICAL BLOCK DIAGRAM OF IMAGE COMPRESSION
Description:
The figure shown above is a schematic block diagram of the system suitable to practice an aspect
of the present methods of compression.
BACKGROUND OF THE IMAGE COMPRESSION:
The present method relates to the digital image processing arts. It finds particular
application in conjunction with processing text symbols in a token based compression system, and
will be described with particular reference thereto. However, it is to be appreciated that the code is
applicable to image output processing of any token based or symbol dependent compression
technique.
Typically image output data streams or bitmaps are optimized for particular printers or
output devices. However, it is known that not all print engines respond identically even when driven
by the same bitmap. The result is that a black and white image on printer A will look somewhat
different than the same image produced by printer B. Technology has been developed that
receives a bitmapped representation of an entire image or page, recognizes it was produced or
generated for a particular printer, and converts or optimizes the bitmap for output on another
printer. Typical conversions include morphological operations such as thickening or thinning lines
and the like. Unfortunately, when compressed files are routed, the image is typically entirely
decompressed, then processed as needed for optimization on a particular output device.
The present technique contemplates a new method and apparatus to process compressed digital
image data which overcomes the above-referenced problems and other.
BRIEF DESCRIPTION OF THE DRAWINGS:
The image compression method may take form in components and various
arrangements of components, and in various steps and arrangements of steps. The drawings are
only for purposes of illustrating the preferred embodiments, and are not to be construed as limiting
the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS:
With reference now to shown figure, an imaging apparatus 10 such as an image
output device, printer, server and the like, receives a stream of image data 12. The data is
preferably compressed according to a token or pattern matching process such as JPEG or other
compression techniques which store single copies of patterns in a document. Those skilled in the
art, however, will appreciate that the teachings here are equally applicable to other pattern-based
substitution processes. Included in the stream of image data 12 are coded image content data 14
or representations and a decoding pattern dictionary 16 or database. The coded image content
data 14 carries a token identifier that points to or represents a particular token in the dictionary 16.
The dictionary 16 contains image patterns substitutable for the coded image content data 14 upon
output processing. For example, a document may be compressed by storing only a single
electronic representation of the letter “a.” The coded image content data 14 includes an identifier
instead of an entire bitmap representation anywhere the letter exists in the document. The identifier
indicates a location in the decoding pattern dictionary 16 containing a single stored instance of an
output-ready representation of the letter. Those skilled in the art will appreciate that in addition to
text patterns for individual letters, similarly groups of characters also repeat frequently enough to
usefully contribute to compression schemes. Likewise, other repeating image characteristics
compressed by a token or pattern matching process will benefit from the teachings of the present
invention.
A decompression processor 20 within the apparatus 10 receives the image data
12. In one embodiment, the processor 20 first identifies and decompresses only the decoding
pattern dictionary 16. Because the decoding pattern dictionary 16 ideally includes only single
instances of tokens or patterns, it will typically be smaller than the coded image content data 14. A
dictionary image processor 22 revises or performs image output processing on individual image
patterns in the decoding pattern dictionary 16. The revision includes conventional image quality
improvements, anti-aliasing, morphological operations such as edge enhancement, dilation,
erosion and others that are readily implemented by those skilled in the art. Alternatively, the
revision processing includes selective gray-scaling, color matching, font substitution and other
image processing which tends to be output device specific. In the illustrated embodiment of the
invention, the apparatus 10 replaces substantially all of the original image patterns in the decoding
pattern dictionary 16 with their respective revised image patterns before the coded content 14 is
decoded.
With continued reference to shown figure, to produce a hard copy output of the
compressed digital image, the apparatus 10 then decompresses or decodes the coded image
content data 14. The coded image content data 14 is parsed and a token identification
corresponding to a location in the decoding pattern dictionary 16 is extracted. The decompression
processor 20 then enters a revised dictionary 24 with the location 26 and returns with the revised
image pattern 28 to be output or otherwise further rendered 30.
[attachment=22822]
Abstract:
Image processing has gone through many years of research and development. Many techniques/
methods have been developed which can be applied to compress images. Compressed Image
processing addresses the problem of reducing the amount of data required to represent an image.
This paper gives the overview of the invention of image compression. Also various methods of
image compression, coding the image data into a compact form are discussed.
An imaging apparatus receives a stream of image data compressed according to a token
based compression algorithm. The token dictionary is decompressed and processed according to
desired standards such as optimizing for a particular image output device. After revision, the
revised tokens are stored in a revised dictionary which is accessed upon decompression of image
content. When the image content is decompressed the locations in the compressed or coded input
data stream refer to locations in the revised dictionary which are extracted and output either to hard
copy or for further processing.
This paper reports some pioneering work in this direction to show that compressed image
processing could well become a new research area and challenge right through the next century
Image Compression
Introduction:
Image compression is minimizing the size in bytes of a graphics file without
degrading the quality of the image to an unacceptable level. The reduction in file size allows more
images to be stored in a given amount of disk or memory space. It also reduces the time required
for images to be sent over the Internet or downloaded from Web pages
Image compression is important for web designers who want to create faster loading web
pages which in turn will make your website more accessible to others. This image compression will
also save you a lot of unnecessary bandwidth by providing high-quality image with fraction of file
size.
Image compression is also important for people who attach photos to emails which will
send the email more quickly, save bandwidth costs and not make the recipient of the email angry.
Sending large image attachments can be considered offensive. This makes people very upset
because the email takes a long time to download and it uses up their precious bandwidth.
Furthermore image compression plays a major role in many important and diverse
applications including televideo conferencing, remote sensing, document and medical imaging,
hazardous waste management application and the like.
LOSSLESS AND LOSSY IMAGE COMPRESSION:
There are several different techniques in which image files can be compressed. A
text file or program can be compressed without the introduction of errors, but only up to a certain
extent. This is called lossless compression. On the other hand, lossy compression reduces a file by
permanently eliminating certain information, especially redundant information.
A typical lossy image compression system is shown in the following figure. There
are three components for the encoder namely Source Encoder, quantizer and entropy encoder.
Source encoder (linear transforms):- A variety of linear transforms have been
developed which include Discrete Fourier Transforms (DFT), discrete cosine transforms (DCT),
discrete wavelet transform (DWT) and many more each with its own advantages and
disadvantages.
Quantizer: - A quantizer simply reduces the number of bits needed to store the
transformed coefficients by reducing the precision of those values. Since this is a many-to-one
mapping, it’s a lossy process and is the main source of compression in an encoder. The
quantization performed on individual coefficient is known as scalar Quantization (SQ), and that
performed on group of coefficients together is called Vector Quantization (VQ).
Entropy encoder: - Entropy encoder further compresses the quantized values
losslessly to give better overall compression. It uses a model to accurately determine the
probabilities for each quantized value and produces an appropriate code based on these
probabilities so that the resultant output code stream will be smaller than the input stream.
PREDICTIVE AND TRANSFORM IMAGE COMPRESSION:
In predictive coding, information already sent are available is used to predict future
values, and the difference is coded. Differential Pulse Code Modulation (DPCM) is an example for
predictive coding.
In transform coding, first the image is transformed from its spatial domain
representation to a different type of representation using some well known transform and then
codes the transformed values (coefficients).
The above figure represents a typical data compression system.
Methods of compressing images:-
RUN LENGTH ENCODING (RLE)
The repeating characters are called runs. Run-Length encoding(RLE) is a very
simple form of data compression in which runs of data are stored as a single data value and count,
rather than as the original run and taking advantage of repetitive data. This is most useful on data
that contains many such runs; for example, some graphic images such as icons and line drawings.
RLE performs lossless data compression and is well suited to palette-based iconic
images. It does not work well at all on continuous-tone images such as photographs. This works
very well for images with solid backgrounds like cartoons
Properties of Image Compression:
There are a few important properties of image compression schemes:
Scalability generally refers to a quality reduction achieved by manipulation of the
bitstream or file (without decompression and re-compression). Other names for scalability are
progressive coding or embedded bitstreams. There are several types of scalability:
Quality progressive or layer progressive: The bitstream successively refines the
reconstructed image.
Resolution progressive: First encode a lower image resolution; then encode the
difference to higher resolutions.
Component progressive: First encode grey; then color.
Region of Interest Coding: Certain parts of the image are encoded with higher
quality than others. This can be combined with scalability (encode these parts first, others later).
Meta Information: Compressed data can contain information about the image which can be
used to categorize, search or browse images. Such information can include color and texture
statistics, small preview images and author/copyright information.
Processing Power: Compression algorithms require different amounts of processing power
to encode and decode. Some high compression algorithms require high processing power.
TYPICAL BLOCK DIAGRAM OF IMAGE COMPRESSION
Description:
The figure shown above is a schematic block diagram of the system suitable to practice an aspect
of the present methods of compression.
BACKGROUND OF THE IMAGE COMPRESSION:
The present method relates to the digital image processing arts. It finds particular
application in conjunction with processing text symbols in a token based compression system, and
will be described with particular reference thereto. However, it is to be appreciated that the code is
applicable to image output processing of any token based or symbol dependent compression
technique.
Typically image output data streams or bitmaps are optimized for particular printers or
output devices. However, it is known that not all print engines respond identically even when driven
by the same bitmap. The result is that a black and white image on printer A will look somewhat
different than the same image produced by printer B. Technology has been developed that
receives a bitmapped representation of an entire image or page, recognizes it was produced or
generated for a particular printer, and converts or optimizes the bitmap for output on another
printer. Typical conversions include morphological operations such as thickening or thinning lines
and the like. Unfortunately, when compressed files are routed, the image is typically entirely
decompressed, then processed as needed for optimization on a particular output device.
The present technique contemplates a new method and apparatus to process compressed digital
image data which overcomes the above-referenced problems and other.
BRIEF DESCRIPTION OF THE DRAWINGS:
The image compression method may take form in components and various
arrangements of components, and in various steps and arrangements of steps. The drawings are
only for purposes of illustrating the preferred embodiments, and are not to be construed as limiting
the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS:
With reference now to shown figure, an imaging apparatus 10 such as an image
output device, printer, server and the like, receives a stream of image data 12. The data is
preferably compressed according to a token or pattern matching process such as JPEG or other
compression techniques which store single copies of patterns in a document. Those skilled in the
art, however, will appreciate that the teachings here are equally applicable to other pattern-based
substitution processes. Included in the stream of image data 12 are coded image content data 14
or representations and a decoding pattern dictionary 16 or database. The coded image content
data 14 carries a token identifier that points to or represents a particular token in the dictionary 16.
The dictionary 16 contains image patterns substitutable for the coded image content data 14 upon
output processing. For example, a document may be compressed by storing only a single
electronic representation of the letter “a.” The coded image content data 14 includes an identifier
instead of an entire bitmap representation anywhere the letter exists in the document. The identifier
indicates a location in the decoding pattern dictionary 16 containing a single stored instance of an
output-ready representation of the letter. Those skilled in the art will appreciate that in addition to
text patterns for individual letters, similarly groups of characters also repeat frequently enough to
usefully contribute to compression schemes. Likewise, other repeating image characteristics
compressed by a token or pattern matching process will benefit from the teachings of the present
invention.
A decompression processor 20 within the apparatus 10 receives the image data
12. In one embodiment, the processor 20 first identifies and decompresses only the decoding
pattern dictionary 16. Because the decoding pattern dictionary 16 ideally includes only single
instances of tokens or patterns, it will typically be smaller than the coded image content data 14. A
dictionary image processor 22 revises or performs image output processing on individual image
patterns in the decoding pattern dictionary 16. The revision includes conventional image quality
improvements, anti-aliasing, morphological operations such as edge enhancement, dilation,
erosion and others that are readily implemented by those skilled in the art. Alternatively, the
revision processing includes selective gray-scaling, color matching, font substitution and other
image processing which tends to be output device specific. In the illustrated embodiment of the
invention, the apparatus 10 replaces substantially all of the original image patterns in the decoding
pattern dictionary 16 with their respective revised image patterns before the coded content 14 is
decoded.
With continued reference to shown figure, to produce a hard copy output of the
compressed digital image, the apparatus 10 then decompresses or decodes the coded image
content data 14. The coded image content data 14 is parsed and a token identification
corresponding to a location in the decoding pattern dictionary 16 is extracted. The decompression
processor 20 then enters a revised dictionary 24 with the location 26 and returns with the revised
image pattern 28 to be output or otherwise further rendered 30.