Embed Size (px)
Citation preview
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [1]
An 8x8 block to focus on:
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [2]
1263 -54 50 101 81 -37 -42 8
-18 -20 12 -41 -125 -87 -8 2
-30 16 10 -50 -35 5 15 5
19 1 2 20 2 -10 2 0
3 6 -7 -13 5 10 3 -1
1 -3 -2 -1 4 3 1 0
-1 -3 -3 -2 -2 0 0 1
0 0 2 -1 -3 -3 -1 0
120 195 154 101 153 166 162 173
131 190 134 101 160 167 164 171
143 171 117 120 176 166 168 172
176 151 105 152 195 167 168 173
202 147 96 160 207 172 167 173
213 133 91 164 207 169 166 172
214 121 101 171 187 149 160 170
204 110 119 183 166 128 154 165
Spatial Domain vs. DCT Domain
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [3]
1263 -54 50 101 81 -37 -42 8
-18 -20 12 -41 -125 -87 -8 2
-30 16 10 -50 -35 5 15 5
19 1 2 20 2 -10 2 0
3 6 -7 -13 5 10 3 -1
1 -3 -2 -1 4 3 1 0
-1 -3 -3 -2 -2 0 0 1
0 0 2 -1 -3 -3 -1 0
Original DCT
16 11 10 16 24 40 51 61
12 12 14 19 26 58 60 55
14 13 16 24 40 57 69 56
14 17 22 29 51 87 80 62
18 22 37 56 68 109 103 77
24 35 55 64 81 104 113 92
49 64 78 87 103 121 120 101
72 92 95 98 112 100 103 99
Using Perceptual Quantization Table?
1264 -55 50 96 72 -40 -51 0
-24 -24 14 -38 -130 -58 0 0
-28 13 16 -48 -40 0 0 0
14 0 0 29 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
round(DCT./Q).*Q
Q
(quantization table)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [4]
Color InterleavingColor Interleaving
– Different components are ordered into Minimum Coding Unit (MCU)– MCUs define repeating interleaving patterns
Minimum coding unit (MCU)MCU1 = {Y00, Y01, Y10, Y11, U00, V00}MCU2 = {Y02, Y03, Y12, Y13, U01, V01}
• • • • • •• • • • • •
• • • • • •• • • • • •
• • •
• • •
• • •
• • •
YY UU VV
UM
CP
EN
EE
408G
Slid
es (c
reat
ed b
y M
.Wu
& R
.Liu
©20
02)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [5]
475 x 330 x 3 = 157 KB luminance
From
Liu
’s E
E330
(Prin
ceto
n)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [6]
RGB ComponentsRGB ComponentsFr
om L
iu’s
EE3
30 (P
rince
ton)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [7]
Y U V (Y Y U V (Y CbCb Cr) ComponentsCr) Components
Assign more bits to Y, less bits to Cb and Cr
From
Liu
’s E
E330
(Prin
ceto
n)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [8]
JPEG Compression (Q=75%)JPEG Compression (Q=75%)
45 KB, compression ration ~ 4:1From
Liu
’s E
E330
(Prin
ceto
n)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [9]
Lossless Coding Part in JPEGLossless Coding Part in JPEG
Differentially encode DC– (lossy part: DC differences are then quantized.)
AC coefficients in one block– Zig-zag scan after quantization for better run-length
save bits in coding consecutive zeros
– Represent each AC run-length using entropy coding use shorter codes for more likely AC run-length symbols
UM
CP
EN
EE
408G
Slid
es (c
reat
ed b
y M
.Wu
& R
.Liu
©20
02)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [10]
LossyLossy Part in JPEGPart in JPEGQuantization (adaptive bit allocation)– Different quantization step size for different coeff. bands– Use same quantization matrix for all blocks in one image– Choose quantization matrix to best suit the image– Different quantization matrices for luminance and color components
Default quantization table– “Generic” over a variety of images
Quality factor “Q”– Scale the quantization table– Medium quality Q = 50% ~ no scaling– High quality Q = 100% ~ unit quantization step size– Poor quality ~ small Q, larger quantization step
visible artifacts like ringing and blokiness
UM
CP
EN
EE
408G
Slid
es (c
reat
ed b
y M
.Wu
& R
.Liu
©20
02)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [11]
Uncompressed (100KB)
JPEG 75% (18KB)
JPEG 50% (12KB)
JPEG 30% (9KB)
JPEG 10% (5KB)
UM
CP
EN
EE
408G
Slid
es (c
reat
ed b
y M
.Wu
& R
.Liu
©20
02)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [12]
JPEG Compression (Q=75% & 30%)JPEG Compression (Q=75% & 30%)
45 KB 22 KBFrom
Liu
’s E
E330
(Prin
ceto
n)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [13]
Y Y CbCb Cr After JPEG (Q=30%)Cr After JPEG (Q=30%)Fr
om L
iu’s
EE3
30 (P
rince
ton)
JPEG Cb JPEG Cr
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [14]
Lossless Coding Part in JPEG: DetailsLossless Coding Part in JPEG: Details
Differentially encode DC– ( SIZE, AMPLITUDE ), with amplitude range in [-2048, 2047]
AC coefficients in one block– Zig-zag scan for better run-length– Represent each AC with a pair of symbols
Symbol-1: ( RUNLENGTH, SIZE ) Huffman coded
Symbol-2: AMPLITUDE Variable length coded
RUNLENGTH ∈ [0,15] # of consecutive zero-valued AC coefficientspreceding the nonzero AC coefficient ∈ [0,15]
SIZE ∈ [0 to 10 in unit of bits] # of bits used to encode AMPLITUDE
AMPLITUDE ∈ in range of [-1023, 1024]
UM
CP
EN
EE
408G
Slid
es (c
reat
ed b
y M
.Wu
& R
.Liu
©20
02)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [15]
An example of JPEG Coding TableAn example of JPEG Coding TableU
MC
P E
NE
E63
1 S
lides
(cre
ated
by
M.W
u ©
2004
)
Table is from slides at Gonzalez/ Woods DIP book website (Chapter 8)
ENEE631 Digital Image Processing (Spring'06) Lec13 – Transf. Coding & JPEG [16]
Summary of TodaySummary of Today’’s Lectures Lecture
JPEG compression standard– Baseline block-DCT based algorithm
lossy part: quantization with different step size for each coeff. Bandlossless part: differential coding, run-length coding, Huffman
Next week: – Subband coding and Wavelet based compression
Exercise 4: Transform coding
Readings– Jain’s book 11.5
Gonzalez’s book 8.5.2, 8.6.2– Wallace’s tutorial paper on JPEG (see course webpage)
UM
CP
EN
EE
631
Slid
es (c
reat
ed b
y M
.Wu
©20
04)
