Slide đa phương tiện chương 3 digital video

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

ng

th

an

co

ng

Digital Video

cu

u

du
o

Tien Pham Van, Dr. rer. nat.
Hanoi University of Science and
Technology

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

.c
om

Agenda

Email:
C9-411 Dai Co Viet str. 1, Hanoi

cu

u

du
o

ng

th

an

co

ng

• Introduction to video basics
• Video data presentation and rendering
• Video compression and communication

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

cu

u

du
o

ng

th

an

co

ng

.c
om

Introduction to Video
Basics

3

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

Types of Video Signals

an

–
–

The primaries can either be RGB or a luminance-chrominance

transformation of them (e.g., YIQ, YUV).
Best color reproduction
Requires more bandwidth and good synchronization of the three
components

ng

–

.c
om

Component video -- each primary is sent as a separate
video signal.

co

•

th

Composite video -- color (chrominance) and luminance
signals are mixed into a single carrier wave.
Some interference between the two signals is inevitable.

•

cu

u

–

du
o

ng

•

S-Video (Separated video, e.g., in S-VHS) -- a compromise
between component analog video and the composite
video. It uses two lines, one for luminance and another
for composite chrominance signal.
4

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Analog Video

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Analog video is represented as a continuous (time varying) signal; Digital video is
represented as a sequence of digital images

PAL (SECAM) Video
625 scan lines per frame, 25
frames per second (40
msec/frame)
Interlaced, each frame is divided
into 2 fields, 312.5 lines/field
Color representation:
Uses YUV color model

co
an

du
o

ng

th

525 scan lines per frame, 30 fps
(33.37 msec/frame).
Interlaced, each frame is divided
into 2 fields, 262.5 lines/field
20 lines reserved for control
information at the beginning of
each field

So a maximum of 485 lines of
visible data

ng

NTSC Video

cu

u

• Laserdisc and S-VHS have
actual resolution of ~420
lines
• Ordinary TV -- ~320 lines
• Each line takes 63.5
microseconds to scan.
Color representation:

• Uses YIQ color model.
5

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:

C9-411 Dai Co Viet str. 1, Hanoi

Frame Rate and Interlacing

The basic idea is quite simple, single still frames are presented at a high
enough rate so that persistence of vision integrates these still frames
into motion.

co

–

.c
om

Persistence of vision: The human eye retains an image for a
fraction of a second after it views the image. This property is
essential to all visual display technologies.

ng

•

an

Motion pictures originally set the frame rate at 16 frames per
second. This was rapidly found to be unacceptable and the
frame rate was increased to 24 frames per second. In Europe,
this was changed to 25 frames per second, as the European
power line frequency is 50 Hz.

When NTSC television standards were introduced, the frame
rate was set at 30 Hz (1/2 the 60 Hz line frequency). Movies
filmed at 24 frames per second are simply converted to 30
frames per second on television broadcasting.

•

cu

u

du
o

ng

th

•

6

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:

C9-411 Dai Co Viet str. 1, Hanoi

Frame Rate and Interlacing

.c
om

Unfortunately, there is no easy way to "put a shutter" in front of a
television broadcast! Therefore, to arrange for two "flashes" per
frame, the flashes are created by interlacing.

•

cu

u

–

du
o

ng

th

an

co

•

For some reason, the brighter the still image presented to the
viewer, the shorter the persistence of vision. So, bright
pictures require more frequent repetition.
If the space between pictures is longer than the period of
persistence of vision -- then the image flickers. Large bright
theater projectors avoid this problem by placing rotating
shutters in front of the image in order to increase the
repetition rate by a factor of 2 (to 48) or three (to 72) without
changing the actual images.

ng

•

With interlacing, the number of "flashes" per frame is two,
and the field rate is double the frame rate. Thus, NTSC
systems have a field rate of 59.94 Hz and PAL/SECAM systems
7
a field rate of 50 Hz.
CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Scanning Video

cu

u

du
o

ng

th

an

co

ng

• Video is obtained via raster scanning, which transforms a
3-D signal p(x, y, t) into a one-dimensional signal s(t) which
t (time)
can be transmitted.
FrameK
• Progressive scanning: left-to-right and top-to-bottom
– Samples in time: frames/sec

Frame2
– Samples along y: lines
Frame1
– Samples along x: pixels
(only for digital video)
• We perceive the images as
continuous, not discrete:
human visual system
performs the interpolation !
• How many frames, lines, and pixels ? Progressive scanning

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

Interlaced Scanning

an
th
ng

A frame

du

o

1
2
3
4
5
6

co

ng

.c
om

• If the frame rate is too slow - > flickering and jagged movements
• Tradeoff between spatial and temporal resolution
– Slow moving objects with high spatial resolution
– Fast moving objects with high frame rate
• Interlaced scanning: scan all even lines, then scan all odd lines.
• A frame is divided into 2 fields (sampled at different time)

M

Even field

1

cu

u

Odd field

2

3

4

5

6

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

th

an

co

ng

R

G

cu

u

du
o

ng

• Three basic colors
R: Red
G: Green
B: Blue
A picture
consists of
three images

.c
om

RGB Color Model

CuuDuongThanCong.com

B

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

YIQ Color Model

u

du
o

ng

th

an

co

ng

.c
om

YIQ color model: used in NTSC color TV
• Y - Luminance containing brightness and detail (monochrome TV)
• To create the Y signal, the red, green and blue inputs to the Y signal
must be balanced to compensate for the color perception misbalance
of the eye.
– Y = 0.3R + 0.59G + 0.11B
• Chrominance
– I = 0.6R – 0.28G - 0.32B (cyan-orange axis)
– Q = 0.21R – 0.52G + 0.31B (purple-green axis)
Y
• Human eyes are most sensitive to Y,
next to I, next to Q.

cu

I

Q

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

YUV Color Model

ng

.c
om

• YUV color model: used for PAL TV and CCIR 601 standard
• Same definition for Y as in YIQ model
• Chrominance is defined by U and V – the color differences

an

co

– U=B–Y
– V=R–Y

du
o

ng

th

Y

cu

u

U

V

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

YCrCb Color Model

du
o

ng

th

an

co

ng

• YCbCr color model: used in JPEG and MPEG
• Closely related to YUV: scaled and shifted YUV
– Cb = ((B – Y)/2) + 0.5
– Cr = ((R – Y)/1.6) + 0.5

cu

u

• Chrominance value in YCbCr are always in the
range of 0 to 1 (normalization)
Make digital processing easy

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Color Models in Video (Cont…)

ng

• Color models based on linear transformation from
RGB color space

cu

u

du
o

ng

th

an

co

C = M3x3 x CRGB

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Analog NTSC and PAL Video
• NTSC Video: Japan, US, …

ng

th

an

co

ng

- 525 scan lines per frame, 30 frames per second
- Interlaced, each frame is divided into 2 fields, 262.5 lines/field
- 20 lines reserved for control information at the beginning of each field
- So a maximum of 485 lines of visible data
- Color representation: YIQ color model

du
o

• PAL Video: China, UK, …

cu

u

- 625 scan lines per frame, 25 frames per second (40 msec/frame)
Interlaced, each frame is divided into 2 fields, 312.5 lines/field
- Uses YUV color model
- Approximately 20% more lines than NTSC
- NTSC vs. PAL
roughly same bandwidth

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Digital Video

Advantages over analog:

th

In a 4:4:4 scheme, each 8×8 matrix of RGB pixels converts to three YCrCb
8×8 matrices: one for luminance (Y) and one for each of the two
chrominance bands (Cr and Cb).
A 4:2:2 scheme also creates one 8×8 luminance matrix but decimates every
two horizontal pixels to create each chrominance-matrix entry. Thus
reducing the amount of data to 2/3rds of a 4:4:4 scheme.

Ratios of 4:2:0 decimate chrominance both horizontally and vertically,
resulting in four Y, one Cr, and one Cb 8×8 matrix for every four 8×8 pixelmatrix sources. This conversion creates half the data required in a 4:4:4
chroma ratio.

–

du
o

ng

–

an

co

ng

Almost all digital video uses component video
The human eye responds more precisely to brightness information than it
does to color, chroma subsampling (decimating) takes advantage of this.

u

•
•

Direct random access --> good for nonlinear video editing
No problem for repeated recording

No need for blanking and sync pulse

.c
om

–
–
–

–

cu

•

Email:
C9-411 Dai Co Viet str. 1, Hanoi

16

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

4:2:2

cu

u

du
o

ng

th

an

co

ng

• Chroma subsampling: human visual
system is more sensitive to
luminance than chrominance
We can subsample chrominance
• 4:4:4 – No subsampling
• 4:2:2, 4:1:1 – horizontally subsample
• 4:2:0 – horizontally and vertically

.c
om

Luma Sampling and Chroma Sub-Sampling

4:1:1

CuuDuongThanCong.com

4:2:0

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Bài tập 1

Email:
C9-411 Dai Co Viet str. 1, Hanoi

cu

u

du
o

ng

th

an

co

ng

.c
om

Cho một ảnh màu kích thước 288x352 pixel, lấy mẫu 4:2:2, giá
trị mỗi điểm ảnh nhận 0…255 (8bit/mẫu).
a/ Tính dung lượng của ảnh
b/ Nén JPEG cho ảnh trên, biết dung lượng ảnh sau nén là 50KB.
Tính tỷ số nén của ảnh?

18

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Bài tập 2

Email:
C9-411 Dai Co Viet str. 1, Hanoi

cu

u

du
o

ng

th

an

co

ng

.c
om

Cho một ảnh màu kích thước 288x352 pixel, lấy mẫu 4:2:2, giá
trị mỗi điểm ảnh nhận 0…255 (8bit/mẫu).
Mã hóa JPEG cho ảnh trên, biết tỷ số nén cho ảnh chói Y là 10
lần, tỷ số nén cho các tín hiệu hiệu màu Cb,Cr là 20 lần.
Tính tỷ số nén của ảnh?

19

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

Sample Quantization – Pixel Resolution

du
o

ng

th

an

co

ng

.c
om

• Pixel resolution depends quantization levels/bits
• Usually, 8 bits for each luma/chroma sample when no compression
8bits/1byte per pixel for gray image, 24bits/3bytes for true color image

cu

u

Luminace (gray) picture
Num. Level Bit
(a)
2
1 (Monochrome)
(b)
4
2
(c)
8
3
(d)
16
4
(e)
32
5
(f)
64
6

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Digital Video

Analog TV is a continuous signal
Digital TV uses discrete numeric values
– Signal is sampled, and samples are quantized
–

160 352

Image represented by pixel array

800

1280

ng

QSIF
(19Kp)

1152

co

120

•

720

Sub-sampling to reduce image resolution or size

.c
om

•
•

Email:
C9-411 Dai Co Viet str. 1, Hanoi

SIF (82Kp)

du
o

486

ng

601 (300Kp)

th

an

240

cu

600

u

SVGA (500Kp)

ATV (1Mp)

720
Workstation (1Mp)

900

HDTV (2Mp)
1080
CuuDuongThanCong.com

/>
1920


Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Freq.
MHz

1050

16:9

P

8

HDTV
Eur, ana

1250

16:9

2.4

P

9

HDTV
NHK

1125

3.3

I

20

NTSC©

525

4:3

7

I

4.2

525

4:3

5

P

4.2

625

4:3

6

I

5.5

625

4:3

4.3

P

5.5

SECAM©

625

4:3

6

I

6

SECAM

625

4:3

4.3

P

6

2.5H

ng

HDTV
USA, ana

Opt.
View
dist

co

Aspect
Ratio

16:9

ng

th

Name

.c

om

P/I

an

Lines

du
o

HDTV

Email:
C9-411 Dai Co Viet str. 1, Hanoi

NTSC

cu

u

PAL©
PAL

© : Conventional

CuuDuongThanCong.com

22

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

Computer Video Format

–

IRIS video board VINO takes NTSC video signal and after digitization can achieve frame
resolution of 640x480 pixels, 8 bits/pixel and 4 fps.
SunVideo digitizer captures NTSC video signal in the form of an RGB signal with frame
resolution of 320x240 pixels, 8 bits/pixel and 30 fps.

ng

–

.c
om

Depends on the i/p and o/p devices (digitizers) for motion video medium.
Digitizers differ in frame resolution, quantization and frame rate

co

•
•

–

The Color Graphics Adapter (CGA):

an

Computer video controller standards

The Enhanced Graphics Adapter (EGA):

ng

–

th

320 x 240 pixels x 2 bits/pixel = 16,000 bytes (storage capacity per image)

–

du
o

640 x 350 pixels x 4 bits/pixel = 112,000 bytes

The Video Graphics Array (VGA):
640 x 480 pixels x 8 bits/pixel = 307,200 bytes

u

The 8514/A Display Adapter Mode:

cu

–

1024 x 768 pixels x 8 bits/pixel = 786,432 bytes

–

The Extended Graphics Array (XGA):
1024x768 at 256 colors or 640x480 at 65,000 colors

–

The Super VGA (SVGS):
Upto 1024x768 pixels x 24 bits/pixel = 2,359,296 bytes
23

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:

C9-411 Dai Co Viet str. 1, Hanoi

cu

u

du
o

ng

th

an

co

ng

.c
om

Video Compression

24

CuuDuongThanCong.com

/>

Pham Van Tien, Dr. rer. nat. , Embedded Networking Research Group
Faculty of Elec. and Telecom, Hanoi University of Science and Technology

Email:
C9-411 Dai Co Viet str. 1, Hanoi

.c
om

Introduction (1/2)

an

co

ng

• Why video compression technique is
important ?
• One movie video without compression

cu

u

du
o

ng

th

– 720 x 480 pixels per frame
– 30 frames per second
– Total 90 minutes
– Full color
– The full data quantity = 167.96 G bytes !!
25

CuuDuongThanCong.com

/>