Digital Signal Processing Reference
In-Depth Information
Table 14.12
The Quality Factor for Luminance
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
Table 14.13
The Quality Factor for Chrominance
17
18
24
47
99
99
99
99
18
21
26
66
99
99
99
99
24
26
56
99
99
99
99
99
47
66
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
number of bits can be used for encoding the DCT coefficients. There are two different quantization
tables, one for luminance (which is the same as the one in the last section and listed here again for
comparison) and the other for chrominance (
Tables 14.12 and 14.13
).
We can see that the chrominance table has numbers with larger values, so that small values of DCT
coefficients will result and hence fewer bits are required for encoding each DCT coefficient. Zigzag
ordering to produce the JPEG vector is similar to the grayscale case, except that there are three JPEG
vectors.
Differential Pulse Code Modulation on Direct-Current Coefficients
Since each 8
8 image block has only one DC coefficient, which can be a very large number and
varies slowly, we make use of DPCM for coding DC coefficients. As an example for the first five image
blocks, the DC coefficients are 200, 190, 180, 160, and 170. DPCM with a coding rule of
dðnÞ¼DCðnÞDCðn
1
Þ
with initial condition
dð
0
Þ¼DCð
0
Þ
produces a DPCM sequence of
200
;
10
;
10
;
20
;
10
Hence, the reduced signal range of these values is feasible for entropy coding.
Run-Length Coding on Alternating-Current Coefficients
The run-length method encodes the pair of
•
the number of zeros to skip and
•
the next nonzero value.
Search WWH ::
Custom Search