Mmdc 2009 Week9
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Mmdc 2009 Week9 as PDF for free.
More details
- Words: 934
- Pages: 45
[1]
Multimedia Data Compression
Presentation parts
Humaid
Introduction
Image Compression
Fan
Video Compression
Mai
Audio Compression
Conclusion
Data Compression
?What
?Why
?How
What is Data Compression?
Why do we need Compression?
[2]
How can we compress data ? Compression Techniques
Audio
Image
Video
MP3
JPEG
MPEG
JPEG Standard Diagram
DC
AC
Original Image Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Color Space Transformation Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
2D-DCT Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Quantization Original Image
Original Image Color Space Transformation
Color Space Transformation
DCT
Quantization Entropy Encoding
DCT
Quantization Entropy Encoding Compressed Image
Compressed Image
Zig-Zag Scanning Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Compressed Image Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
RGB Histograms O rig in a lR e d C o m p o n e n t
T h e R e d C o m p o n e n t
R e c o n s tru c te d R e d C o m p o n e n t
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
O rig in a lG re e n C o m p o n e n t
T h e G re e n C o m p o n e n t
3 0 0 0
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0
1 5 0
2 0 0
2 5 0
2 0 0
2 5 0
2 0 0
2 5 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
O rig in a lB lu e C o m p o n e n t
T h e B lu e C o m p o n e n t
5 0
1 0 0
1 5 0
R e c in s tru c te d B lu e C o m p o n e n t
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0 0
1 0 0
R e c o n s tru c te d G re a n C o m p o n e n t
3 0 0 0
0
5 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
5 0
1 0 0
1 5 0
Images Mesh Plots
Compression Performance Tests • The Original Image Size
= 718,149 KBytes
• The Compressed Image Size = 147,905 KBytes • The compression ratio
= 4.86
• The Mean Square Error
= 32.07
• The Processing Time is
= 2.5 s
Video Compression
Video compression Video compression
Compression standard
Compression processing
Video to frames
Frame1 Frame2
Frame3
Frame4
Frame • Slice • Macro block
Video processing
YCbCr color model
8X8 2D-DCT
2d-DCT
DCT2 of Original Image - Red
4
3
x 10
2.5
2
1.5
1
0.5
0
-0.5
0
20
40
60
80
100
120
140
160
180
2d-Dct
4
3
x 10
DCT2 of Original Image - Red
2.5
2
1.5
1
0.5
0
-0.5
0
50
100
150
Quantization
Quantization
Zig-Zag
Entropy Encoding
Audio Compression
MP3 Diagram • Following is the main diagram for Mp3 encoder
MPEG-1 Layer 3 or MP3 • MP3 is a digital audio encoding format using a form of lossy data compression. • The main step in the MP3 compression process is the Modified Discrete Cosine Transform or MDCT. • It has a %50 overlapping feature. •
The output of MDCT is N/2 for every N input.
Implementation of MDCT • This part of the project was concentrating on the implementation of MDCT in MATLAB. • Studying its effect on compressing audio samples.
MDCT CR=3.3 2 (2) 1 0 -1 -2
0
100
200
300 400 Time, (1) The MDCT
500
600
2 (3) 1 0 -1 -2
0
100
200 300 400 Time, (2)The Compressed MDCT
500
600
Original & Reconstructed Signal e = 0.0965 1 (1) 0.5 0 -0.5 -1 0
0.5
1
1.5 2 2.5 Time, (1)Theoriginal Signal
3
3.5
4 x10
4
1 (2) 0.5 0 -0.5 -1 0
0.5
1
1.5 2 2.5 Time, (2)final output Signal
3
3.5
4 x10
4
MDCT CR=2 2 (2) 1 0 -1 -2 0
100
200
300 Time, (1)TheMDCT
400
500
600
2 (3) 1 0 -1 -2 0
100
200 300 400 Time, (2)TheCompressedMDCT
500
600
Original & Reconstructed Signal e = 0.1275 1 (1 ) 0 .5 0 -0 .5 -1 0
0 .5
1
1 .5 2 T im e ,(1 )T h e o rig in a lS ig n a l
2 .5
3
3 .5
4 x1 0
4
1 (2 ) 0 .5 0 -0 .5 -1 0
0 .5
1
1 .5 2 T im e ,(2 )fin a lo u tp u tS ig n a l
2 .5
3
3 .5
4 x1 0
4
Conclusion
This Semester Project Plan Implementing Current industrial Compression Algorithms in MATLAB.
Project Phases
Semester 2 2008
Semester 1 2009
Semester 2 2009
Algorithms Research
Implementation by MATLAB
Implementation by Simulink
The End
Questions
References [1]
http://plus.maths.org/issue23/features/data/data.jpg [2] [3]
http://leadinganswers.typepad.com/photos/uncategorized/2008/ 01/27/computer_users.jpg http://www.researchandtechnology.net/pcif/filtering_example.p hp
[4] http://ugweb.cs.ualberta.ca/~c301/F07/labs/images/rgb.gif http://webstyleguide.com/graphics/graphics/7.03.gif
References [1] wikipedia website [2] http://www.mp3-converter.com [3] http://www.mp3-download.be/how-mp3compression-works.php
Multimedia Data Compression
Presentation parts
Humaid
Introduction
Image Compression
Fan
Video Compression
Mai
Audio Compression
Conclusion
Data Compression
?What
?Why
?How
What is Data Compression?
Why do we need Compression?
[2]
How can we compress data ? Compression Techniques
Audio
Image
Video
MP3
JPEG
MPEG
JPEG Standard Diagram
DC
AC
Original Image Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Color Space Transformation Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
2D-DCT Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Quantization Original Image
Original Image Color Space Transformation
Color Space Transformation
DCT
Quantization Entropy Encoding
DCT
Quantization Entropy Encoding Compressed Image
Compressed Image
Zig-Zag Scanning Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
Compressed Image Original Image Color Space Transformation DCT
Quantization Entropy Encoding Compressed Image
RGB Histograms O rig in a lR e d C o m p o n e n t
T h e R e d C o m p o n e n t
R e c o n s tru c te d R e d C o m p o n e n t
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
O rig in a lG re e n C o m p o n e n t
T h e G re e n C o m p o n e n t
3 0 0 0
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0
1 5 0
2 0 0
2 5 0
2 0 0
2 5 0
2 0 0
2 5 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
O rig in a lB lu e C o m p o n e n t
T h e B lu e C o m p o n e n t
5 0
1 0 0
1 5 0
R e c in s tru c te d B lu e C o m p o n e n t
2 0 0 0
2 0 0 0
1 0 0 0
1 0 0 0
0 0
1 0 0
R e c o n s tru c te d G re a n C o m p o n e n t
3 0 0 0
0
5 0
0 5 0
1 0 0
1 5 0
2 0 0
2 5 0
0
5 0
1 0 0
1 5 0
Images Mesh Plots
Compression Performance Tests • The Original Image Size
= 718,149 KBytes
• The Compressed Image Size = 147,905 KBytes • The compression ratio
= 4.86
• The Mean Square Error
= 32.07
• The Processing Time is
= 2.5 s
Video Compression
Video compression Video compression
Compression standard
Compression processing
Video to frames
Frame1 Frame2
Frame3
Frame4
Frame • Slice • Macro block
Video processing
YCbCr color model
8X8 2D-DCT
2d-DCT
DCT2 of Original Image - Red
4
3
x 10
2.5
2
1.5
1
0.5
0
-0.5
0
20
40
60
80
100
120
140
160
180
2d-Dct
4
3
x 10
DCT2 of Original Image - Red
2.5
2
1.5
1
0.5
0
-0.5
0
50
100
150
Quantization
Quantization
Zig-Zag
Entropy Encoding
Audio Compression
MP3 Diagram • Following is the main diagram for Mp3 encoder
MPEG-1 Layer 3 or MP3 • MP3 is a digital audio encoding format using a form of lossy data compression. • The main step in the MP3 compression process is the Modified Discrete Cosine Transform or MDCT. • It has a %50 overlapping feature. •
The output of MDCT is N/2 for every N input.
Implementation of MDCT • This part of the project was concentrating on the implementation of MDCT in MATLAB. • Studying its effect on compressing audio samples.
MDCT CR=3.3 2 (2) 1 0 -1 -2
0
100
200
300 400 Time, (1) The MDCT
500
600
2 (3) 1 0 -1 -2
0
100
200 300 400 Time, (2)The Compressed MDCT
500
600
Original & Reconstructed Signal e = 0.0965 1 (1) 0.5 0 -0.5 -1 0
0.5
1
1.5 2 2.5 Time, (1)Theoriginal Signal
3
3.5
4 x10
4
1 (2) 0.5 0 -0.5 -1 0
0.5
1
1.5 2 2.5 Time, (2)final output Signal
3
3.5
4 x10
4
MDCT CR=2 2 (2) 1 0 -1 -2 0
100
200
300 Time, (1)TheMDCT
400
500
600
2 (3) 1 0 -1 -2 0
100
200 300 400 Time, (2)TheCompressedMDCT
500
600
Original & Reconstructed Signal e = 0.1275 1 (1 ) 0 .5 0 -0 .5 -1 0
0 .5
1
1 .5 2 T im e ,(1 )T h e o rig in a lS ig n a l
2 .5
3
3 .5
4 x1 0
4
1 (2 ) 0 .5 0 -0 .5 -1 0
0 .5
1
1 .5 2 T im e ,(2 )fin a lo u tp u tS ig n a l
2 .5
3
3 .5
4 x1 0
4
Conclusion
This Semester Project Plan Implementing Current industrial Compression Algorithms in MATLAB.
Project Phases
Semester 2 2008
Semester 1 2009
Semester 2 2009
Algorithms Research
Implementation by MATLAB
Implementation by Simulink
The End
Questions
References [1]
http://plus.maths.org/issue23/features/data/data.jpg [2] [3]
http://leadinganswers.typepad.com/photos/uncategorized/2008/ 01/27/computer_users.jpg http://www.researchandtechnology.net/pcif/filtering_example.p hp
[4] http://ugweb.cs.ualberta.ca/~c301/F07/labs/images/rgb.gif http://webstyleguide.com/graphics/graphics/7.03.gif
References [1] wikipedia website [2] http://www.mp3-converter.com [3] http://www.mp3-download.be/how-mp3compression-works.php
Related Documents
Mmdc 2009 Week9
July 2020 2
Week9.pdf
December 2019 19
Cad Week9
November 2019 19
Week9 Hometown Brochure
November 2019 20
Week9 Homework Scribd
November 2019 6