Quantization Table

Report
Advanced JPEG
Dinesh Ganotra
Organizations Involved




Joint Photographic Experts Group
ISO International Organization for Standards
CCITT International Telegraph and Telephone
Consultative Committee
IEC International Electro-technical Commission
Recap
Contrast Sensitivity of the Eye
◦ At a 1 meter viewing
distance an ideal digital
image 40cm wide should
have about 4,000 pixels.
◦ The contrast sensitivity
curve for luminance is
different from
chrominance.
QUANTIZATION TABLE
Luminance Quantization Table
Chrominance Quantization Table
ZERO SHIFT
The 8-bit image data is zero-shifted by subtracting 128 from the
data before the DCT is calculated. For 12-bit data the zero shift is
accomplished by subtracting 2048. This reduces the internal
precision requirement for DCT calculation.
DCT
One multiplication and nine
additions per coefficient
×+
+
+
+
+
+
+
+
+
Entropy
PCM
600
500
DPCM
400
300
200
101
100
11
10
0
0
001
100
Huffman
Coding
Arithmatic
Coding
DPCM
lossless
Sequential
DCT
lossy
Progressive
Huffman
Arithmetic
Hierarchical
Huffman
Computationally Simpler
Requires code tables to be known
Arithmetic
Higher Performance ( 10%)
code book adapts dynamically
to the data being coded.
Sequential
codes the image in a single scan or passes
through the data.
The decoder necessarily follows the same
order in decoding the data
Progressive
encodes the image in two or more scans
or passes through the data, first coding
an approximation of the entire image
with each succeeding scan.
decoder follows the same order in decoding,
first reconstructing the approximating of the
image, and then adding finer detail with each
succeeding scan
AC band 1
AC band 2
AC band 3
DC band 1
magnitude bit plane
Progressive
2
1
0
AC band 4
DC band 4
AC band 5
AC band 6
Hierarchical
progressive coding with increasing spatial
resolution between progressive stages
Data Unit
Y1
Y2
Y3
Y4
Cb1
Cb2
Cr1
Cr2
Y5
Y6
Y7
Y8
Cb3
Cb4
Cr3
Cr4
Horizontally interleaved data units
Vertically interleaved data units
Y1
Y2
Cb1
Cb2
Cr1
Cr2
Y3
Y4
Cb3
Cb4
Cr3
Cr4
Y5
Y6
Y7
Y8
Non interleaved, horizontal , and
vertical interleaved
Scan1 : Y1 , Y2 , Y3 , Y4 , Y5 , Y6 , Y7 , Y8 Scan2 : Cb1 , Cb2 , Cb3 , Cb4 Scan3 : Cr1 , Cr2 , Cr3 , Cr4
Scan1 : Y1 , Y2 , Cb1 , Cr1, Y3 , Y4 , Cb2 , Cr2, Y5 , Y6 , Cb3 , Cr3 , Y7 , Y8 , Cb4 , Cr4
Scan1 : Y1 , Y3 , Cb1 , Cr1, , Y2 , Y4 , Y5 , Y7 , Cb3 , Cr3 , Y6 , Y8 , Y7 , Y8 , Cb4 , Cr4
Example
FFD8
FFC9
00 11 08 00 80 00 80 03 01 11 00 02
11 00 03 11 00
FFCC
00 06 00 10 10 05
FFDA
00 08 01 01 01 00 00 3F 00 …
FFCC
00 06 01 10 11 05
FFDA
00 08 01 02 11 00 3F 00 …
FFDA
00 08 01 03 11 00 3F 00 …
FFD9
Markers
Code
FF01
FF02-FFBF
FFC0
FFC1
FFC2
FFC3
FFC4
FFC5
FFC6
FFC7
FFC8
FFC9
FFCA
FFCB
Name
TEM
RES
SOF0
SOF1
SOF2
SOF3
DHT
SOF5
SOF6
SOF7
JPG
SOF9
SOF10
SOF11
Category
For temporary use in arithmetic coding
Reserved
Nondifferential Huffman –coding frames
Nondifferential Huffman –coding frames
Nondifferential Huffman –coding frames
Nondifferential Huffman –coding frames
Define Huffman Table(s)
Differential Huffman –coding frames
Differential Huffman –coding frames
Differential Huffman –coding frames
Reserved for JPEG extension
Nondifferential arithmetic–coding frames
Nondifferential arithmetic–coding frames
Nondifferential arithmetic–coding frames
Baseline DCT
Extended sequential DCT
Progressive DCT
Lossless sequential
differential sequential DCT
differential progressive DCT
differential lossless
Extended sequential DCT
Progressive DCT
Lossless sequential
Markers Cont..
FFCC
FFCD
FFCE
FFCF
FFD0-FFD7
FFD8
FFD9
FFDA
FFDB
FFDC
FFDD
FFDE
FFDF
FFE0-FFEF
FFF0-FFFD
FFFE
DAC
Define Arithmatic Coding
SOF13
Differential arithmetic- coding frames
differential sequential
SOF14
Differential arithmetic- coding frames
differential progressive
SOF15
Differential arithmetic- coding frames
differential lossless
RSTm Restart with modulo 8 counter m
SOI
Start of image
EOI
End of Image
SOS
Start of Scan
DQT
Define quantization table
DNL
Define number of lines
DRI
Define restart interval
DHP
Define Hierarchical progression
EXP
Expand reference images
APPn
Reserved for application use (two different applications may have conflicting usage of APPn)
JPGn
Reserved for JPEG extension
COM
Comment
Frames and Scans
Frame Header SOFn
Parameter
Marker ( FFC0-3, 5-7, 9-B,D-F)
Frame header length
Sample precision
Number of lines
Number of samples per line
Number of components in frame
Frame component specification
Component identifier
Horizontal sampling factor
Vertical sampling factor
Quantization table destination selector
Nf
03
C1
00
H1
2
V1
1
Tq
00
C2
01
H2
11
V2
01
Symbol
SOFn
Lf
P
Y
X
Nf
Size
16
16
8
16
16
8
Ci
Hi
Vi
Tqi
8
4
4
8
Tq2
02
C3
02
H3
1
V3
1
Tq3
01
The above example shows a three component system.YCrCb with identical sampling for all components in
the vertical direction and twice as many Y samples as Cr or Cb samples in the horizontal direction. Let the
first component use quantization table 0 and the rest use quantization table 1. If Y, Cr and Cb are numbered
0, 1, and 2 respectively, the components specification is as given above.
Other Header Structures
Scan Header
DHT Marker Segment
DAC Marker Segment
DQT Marker Segment
Restart Segment
References
JPEG still image data compression standard,
William B Pennebaker, JL Mitchell,
VNR , International Thomson Publishing Inc.

similar documents