jpg image processing nagham salim_as.ppt

JPEG
Nagham ALLayla & Asmaa ALQassab
Supervised by
Dr. Khalil ALSaif

What is JPEG?
 JPEG is sophisticated lossy/lossless
compression method for color or greyscale
still images. It works best on continuous tone
images where adjacent pixels have similar
colors. One advantage of JPEG is the use of
many parameters, allowing the use to adjust
the amount of data lost over a wide variety
range.

Main Goals of JPEG
 High Compression Ratio in such cases where
images quality is judged as very good to
excellent.
 The use of many parameters allowing user to
experiment and achieve the desired level of
compression.
 Obtaining good results any kind of continuous
tone images.
 Sophisticated but not too complex compression
method, allowing software and hardware
implementation on many platform.
 Several modes of operations.

Modes of JPEG
 Sequential Mode: each image component is
compressed in a single left to right, top to bottom
scan.
 Progressive Mode: the images is compressed in
multiple blocks to be viewed from coarse to final
detail.
 Lossless Mode: important for cases where the
user decides no pixel should be lost.
 Hierarchical Mode: the image is compressed at
multiple resolution allowing lower resolution blocks
to be viewed without first having to decompressed
the following higher-resolution blocks.

Basic JPEG Compression
 JPEG compression involves the following:
 Color Space Transform and subsampling (YIQ)
 DCT (Discrete Cosine Transformation)
 Quantization
 Zigzag Scan
 DPCM on DC component & RLE on AC Components
 Entropy Coding — Huffman or Arithmetic

1.Colour Space Transform and sub-sampling(YIQ)
 The images values are transformed from RGB to
luminance/chrominance color space.
 The eye is sensitive to small changes in luminance
but not in chrominance. So chrominance values
can be used to compress the data.
 This step is optional but important since the
remainder of the algorithm works on each color
component separately.

Color Space Transformation
 RGB to YCbCr
 Y is Luminance (Brightness).
 CbCr is the Chrominance.
 Sampling:
RGB
RGB
Y
Cb Cr
+ (Down Sampling)

RGB to YCbCr
 First the image is divided into 8x8 blocks.
RGB RGB
RGB RGB
Y Y
Y Y
Cb
Cr
+
8
8
8 8
8
8
8 8
8
8
8
4 x 64 x 3 byte 4 x 64 x 1 byte 2 x 64 x 1 byte
Y = 0.299 R + 0.587 G + 0.114 B
Cb = - 0.169 R - 0.331 G + 0.500 B
Cr = 0.500 R - 0.419 G - 0.081 B

The mathematical representation of FDCT (2-D):

 



7
0
7
0
)
16
/
)
1
(
2
cos(
)
16
/
)
1
(
2
cos(
)
,
(
)
(
)
(
4
1
)
,
(
x y
v
j
u
i
y
x
f
v
C
u
C
v
u
F 

Where


 

otherwise
x
x
C
1
0
2
/
1
)
(
f(x,y): 2-D sample value
F(u,v): 2-D DCT coefficient
2.Discrete Cosine Transform (DCT)
 The Discrete Cosine Transform (DCT) separates the frequencies
contained in an image.
 The original data could be reconstructed by Inverse DCT.

DCT (cont’d)
DCT
Coefficients
sxy DCT
8
8 8
8
DC coefficient
 The DCT coefficient values can be regarded as the
relative amounts of the 2-D spatial frequencies
contained in the 88 block
 the upper-left corner coefficient is called the DC
coefficient, which is a measure of the average of
the energy of the block
 Other coefficients are called AC coefficients,
coefficients correspond to high frequencies tend to
be zero or near zero for most natural images

DCT (cont’d)
 Why we apply DCT?
 To move from Spatial Domain to Frequency
Domain.
 Why not FFT?
 DCT is like FFT, but can approximate linear
signals well with few [real] coefficients.
 DCT is slow
 1024 multiplications and 896 additions.
 JPEG 2000: Wavelet Transforms.

3.Quantization
)
)
,
(
)
,
(
(
)
,
(
'
v
u
Q
v
u
F
Round
v
u
F 
F(u,v): original DCT coefficient
F’(u,v): DCT coefficient after quantization
Q(u,v): quantization value

Quantization (cont’d)
 Why do we need to quantise:
 To throw out bits from DCT.
Example: 101101 = 45 (6 bits).
Truncate to 4 bits: 1011 = 11.
Truncate to 3 bits: 101 = 5.
 Quantization error is the main source of Lossy
Compression.
 • DCT itself is not Lossy
 • How we throw away bits in Quantization
Step is Lossy

Quantization (Example)
F(u,v)
8x8 DCT coefficiences
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
Q(u,v)
Quantization matrix
699.25 43.18 55.25 72.11 24.00 -25.51 11.21 -4.14
-129.78 -71.50 -70.26 -73.35 59.43 -24.02 22.61 -2.05
85.71 30.32 61.78 44.87 14.84 17.35 15.51 -13.19
-40.81 10.17 -17.53 -55.81 30.50 -2.28 -21.00 -1.26
-157.50 -49.39 13.27 -1.78 -8.75 22.47 -8.47 -9.23
92.49 -9.03 45.72 -48.13 -58.51 -9.01 -28.54 10.38
-53.09 -62.97 -3.49 -19.62 56.09 -2.25 -3.28 11.91
-20.54 -55.90 -20.59 -18.19 -26.58 -27.07 8.47 0.31

Quantization (Example)
43.70 3.93 5.52 4.51 1.00 -0.64 0.22 -0.07
-10.82 -5.96 -5.02 -3.86 2.29 -0.41 0.38 -0.04
6.12 2.33 3.86 1.87 0.37 0.30 0.22 -0.24
-2.91 0.60 -0.80 -1.92 0.60 -0.03 -0.26 -0.02
-8.75 -2.25 0.36 -0.03 -0.13 0.21 -0.08 -0.12
3.85 -0.26 0.83 -0.75 -0.72 -0.09 -0.25 0.11
-1.08 -0.98 -0.04 -0.23 0.54 -0.02 -0.03 0.12
-0.29 -0.61 -0.22 -0.19 -0.24 -0.27 0.08 0.00
( , )
( , )
F u v
Q u v
44 4 6 5 1 -1 0 0
-11 -6 -5 -4 2 0 0 0
6 2 4 2 0 0 0 0
-3 1 -1 -2 1 0 0 0
-9 -2 0 0 0 0 0 0
4 0 1 -1 -1 0 0 0
-1 -1 0 0 1 0 0 0
0 -1 0 0 0 0 0 0
( , )
( , )
( , )
q
F u v
F u v
Round
Q u v

 
 
 

4.Zig Zag Scan
 8x8 Blocks to a 1x64
Vector.
 Why Zig Zag Scan?
 To sort 8x8 DCT
coefficients based on
the spatial frequencies
(low frequency
coefficients in top of
vector).
0 1 2 . . . 63
DC Component
(No Frequency)
Highest Frequency
1x64 Vector:
0 1 5 6 14 15 27 28
2 4 7 13 16 26 29 42
3 8 12 17 25 30 41 43
9 11 18 24 31 40 44 53
10 19 23 32 39 45 52 54
20 22 33 38 46 51 55 60
21 34 37 47 50 56 59 61
35 36 49 48 57 58 62 63
Horizontal frequency
Vertical
frequency

Zig-Zag Reordering (Example)
44 4 6 5 1 -1 0 0
-11 -6 -5 -4 2 0 0 0
6 2 4 2 0 0 0 0
-3 1 -1 -2 1 0 0 0
-9 -2 0 0 0 0 0 0
4 0 1 -1 -1 0 0 0
-1 -1 0 0 1 0 0 0
0 -1 0 0 0 0 0 0
( , )
q
F u v
Zig-Zag Reordering :
44,
4,-11,
6,-6,6,
5,-5,2,-3,
-9,1,4,-4,1,
-1,2,2,-1,-2,4,
-1,0,0,-2,0,0,0,
0,0,0,1,0,1,-1,0,
-1,0,-1,0,0,0,0,
0,0,0,-1,0,0,
0,1,0,0,0,
0,0,0,0,
0,0,0,
0,0,
0

5.DPCM & RLE
1. DPCM on DC Component
 Differential Pulse Code Modulation.
 DC component is large and varied, but often
close to previous value.
2. RLE on AC Components
 Zero Run Length Encoding.
 1 x 64 vector has lots of zeros in it.
 57,45,0,0,0,0,23,0,-30,-16,0,0,1,0,0,0, 0 , 0 ,0 , only 0,..,0
RLC: (0,57); (0,45); (4,23); (1,-30); (0,-16); (2,1); EOB
EOB = (0,0)

Differential Pulse Code Modulation(DPCM) on DC
Component
DCi1 DCi
…
…
DC1
DC0
0
…
…
Encode :
Diffi = DCi  DCi1
Decode :
DCi = DCi1 + Diffi
DCs of 88 blocks
block 1 block i1 block i
Diff1=DC1
Diffi1=
DCi1 DCi2
Diffi=
DCi DCi1

6.Huffman (Entropy) Coding
 DC and AC components finally need to be
represented by a
 smaller number of bits (Arithmetic coding also
supported in place of Huffman coding):
 (Variant of) Huffman coding: Each DPCM-coded DC
coefficient is represented by a pair of symbols : (Size,
Amplitude) where Size indicates number of bits needed
to represent coefficient and Amplitude contains actual
bits.

Huffman Coding (Table of values and bits for the
values)
Category Values Bits for the value
1 -1,1 0,1
2 -3,-2,2,3 00,01,10,11
3 -7,-6,-5,-4,4,5,6,7 000,001,010,011,100,101,110,111
4 -15,...,-8,8,...,15 0000,...,0111,1000,...,1111
5 -31,...,-16,16,...31 00000,...,01111,10000,...,11111
6 -63,...,-32,32,...63 000000,...,011111,100000,...,111111
7 -127,...,-64,64,...,127 0000000,...,0111111,1000000,...,1111111
8 -255,..,-128,128,..,255 ...
9 -511,..,-256,256,..,511 ...
10 -1023,..,-512,512,..,1023 ...
11 -2047,..,-1024,1024,..,2047 ...

Huffman Coding (Huffman table of luminance AC
coefficience)
run/category code length code word
0/0 4 1010
...
0/6 7 1111000
...
0/10 16 1111111110000011
1/1 4 1100
...
1/5 11 11111110110
...
4/5 16 1111111110011000
...
15/10 16 1111111111111110

Huffman Coding (Huffman table of luminance DC
coefficience)
category code length code word
0 2 00
1 3 010
2 3 011
3 3 100
4 3 101
5 3 110
6 4 1110
7 5 11110
8 6 111110
9 7 1111110
10 8 11111110
11 9 111111110

Huffman Coding (Example)
Run length coding of 63 AC coefficiences :
(0,57) ; (0,45) ; (4,23) ; (1,-30) ; (0,-8) ; (2,1) ; (0,0)
Encode the right value of these pair as category and bits for the value,
except the special markers like (0,0) or (15,0) :
(0,6,111001) ; (0,6,101101) ; (4,5,10111);
(1,5,00001) ; (0,4,0111) ; (2,1,1) ; (0,0)
The difference of DC coefficience :
-511
Encode the value as category and bits for the value :
9, 000000000

Huffman Coding (Example) (cont’d)
The AC coefficiences :
(0,6,111001) ; (0,6,101101) ; (4,5,10111);
(1,5,00001) ; (0,4,0111) ; (2,1,1) ; (0,0)
Encode the left two value in () using Huffman encoding :
1111000 111001 , 1111000 101101 , 1111111110011000 10111 ,
11111110110 00001 , 1011 0111 , 11100 1 , 1010
The DC coefficience :
9, 000000000
Encode the category using Huffman encoding :
1111110 000000000

jpg image processing nagham salim_as.ppt

More Related Content

Similar to jpg image processing nagham salim_as.ppt

More from naghamallella

Recently uploaded

jpg image processing nagham salim_as.ppt