See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/261653610

A Real-Time H.264/AVC Encoder & Decoder with Vertical Mode for Intra Frame
and Three Step Search Algorithm for P-Frame

Conference Paper · April 2014

DOI: 10.5121/csit.2014.4404

CITATIONS READS

0 1,263

2 authors:

Mohammed Al-Jammas Noor N. Hamdoon

Ninevah University University of Mousl
17 PUBLICATIONS 32 CITATIONS 5 PUBLICATIONS 3 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

PID Controller Configuration and Tuning Based on Genetic Algorithms View project

Investigation and Enhancement of 5G channel coding View project

All content following this page was uploaded by Mohammed Al-Jammas on 16 April 2014.

The user has requested enhancement of the downloaded file.

A REAL-TIME H.264/AVC ENCODER&DECODER
WITH VERTICAL MODE FOR INTRA FRAME AND
THREE STEP SEARCH ALGORITHM FOR P-FRAME

Dr. Mohammed H. Al-Jammas1 and Mrs. Noor N. Hamdoon 2

1
Deputy Dean/College of Electronics Eng./University of Mosul/Mosul-Iraq
[email protected]
2
Department of Electrical Engineering /College of Engineering/Mosul–Iraq
[email protected]

ABSTRACT
The video coding standards are being developed to satisfy the requirements of applications for various
purposes, better picture quality, higher coding efficiency, and more error robustness. The new
international video coding standard H.264 /AVC aims at having significant improvements in coding
efficiency, and error robustness in comparison with the previous standards such as MPEG-2, H261,
H263,and H264. Video stream needs to be processed from several steps in order to encode and decode
the video such that it is compressed efficiently with available limited resources of hardware and
software. All advantages and disadvantages of available algorithms should be known to implement a
codec to accomplish final requirement. The purpose of this project is to implement all basic building
blocks of H.264 video encoder and decoder. The significance of the project is the inclusion of all
components required to encode and decode a video in MatLab .

KEYWORDS
H264/AVC , Intra frame (I-frame) , Inter frame (P-frame)

1. INTRODUCTION
A Digital video compression is an important techniques that enables efficient transmission
bandwidth and storage space of multimedia. The H.264/AVC is a standard video coding that
developed to achieve significant improvements, in the compression performance, over the
existing standards. In fact, the high compression performance comes mainly from the
prediction techniques that remove spatial and temporal redundancies. To remove spatial
redundancy, H.264/AVC intra prediction supports many prediction modes to make better
prediction. Inter prediction is enhanced by motion estimation (ME) to remove more temporal
redundancy. However, the H.264/AVC coding performance comes at the price of
computational complexity[1].
H.264/AVC intra encoding achieve higher compression ratio and picture quality compared
with the latest still image coding standard JPEG2000].intra prediction is the first process of
advanced video coding standard. It predicts a macro block by referring to its previous macro
blocks to reduce spatial redundancy. Intra prediction supports nine modes for 4x4 block and
four modes for 16x16 blocks[2].
H.264 is an open, licensed standard that supports the most efficient video compression
techniques available today. Without compromising image quality, an H.264 encoder can
reduce the size of a digital video file by more than 80% compared with the Motion JPEG
format and as much as 50% more than with the MPEG-4 Part 2 standard. This means that
much less network bandwidth and storage space are required for a video file. Or seen another
way, much higher video quality can be achieved for a given bit rate[3].
And also entered as an adjunct of the kind of evolutionary in public services such as video
storage on the Internet and telecommunications companies and surveillance cameras used in
industrial plants , and due to accept this kind of decoder over a wide range of frames during
the second (60/30/25 (fps)) has been expanding in applications control of highways, airports ,
Most of the controversy over the techniques used to process coded information is how fast
and accurate images and video after the code process it possible to re-information fully if it is
reduced to half?. Will be answered by this research, which includes the representation of the
coded H264 and decode of the file with a high level using the MATLAB simulation software
to achieve complete system of coded data and return it in the same efficient used for the
original system.

2. THE ENCODING P ROCESS

H.264 encoder works on the same principles as that of any other codec. Figures (eg, Figure 1)
shows the basic building blocks of H.264 video codec.
The input to the encoder is generally an intermediate format stream, which goes through the
prediction block; the prediction block will perform intra and inter prediction (motion
estimation and compensation) and exploit the redundancies that exist within the frame and
between successive frames. The output of the prediction block is then transformed and
quantized. An integer approximate of the discrete cosine transform is used (DCT) for
transformation. It uses 4x4 or 8x8 integer transform, and outputs a set of coefficients each of
which is a weighted value for a standard basis pattern. The coefficients are then quantized i.e.
each coefficient is divided by an integer value. Quantization reduces the precision of the
transform coefficients according to the quantization parameter (QP). Typically, the result is a
block in which most or all of the coefficients are zero, with a few non-zero coefficients. Next,
the coefficients are encoded into a bit stream. The video coding process creates a number of
parameters that must be encoded to form a compressed bit stream[4]. These values include:
 Quantized transform coefficients.
 Information to re-create prediction.
 Information about the structure of compressed data and the compression tools used
under encoding.
These parameters are converted into binary codes using variable length coding or arithmetic
coding. Each of these encoding methods produces an efficient, compact binary representation
of the information. The encoded bit stream can now be transmitted or stored.

Figure 1. Encode and Decode circuit

3. VIDEO COMPRESSION WORKS
Video compression is about reducing and removing redundant video data so that a digital
video file can be effectively sent and stored. The process involves applying an algorithm to
the source video to create a compressed file that is ready for transmission or storage. To play
the compressed file, an inverse algorithm is applied to produce a video that shows virtually
the same content as the original source video. The time it takes to compress, send,
decompress and display a file is called latency. The more advanced the compression
algorithm, the higher the latency, given the same processing power. A pair of algorithms that
works together is called a video codec (encoder/decoder).
Video codec’s that implement different standards are normally not compatible with each
other; that is, video content that is compressed using one standard cannot be decompressed
with a different standard. For instance, an MPEG-4 Part 2 decoder will not work with an
H.264 encoder. This is simply because one algorithm cannot correctly decode the output from
another algorithm but it is possible to implement many different algorithms in the same
software or hardware, which would then enable multiple formats to be compressed.
Different video compression standards utilize different methods of reducing data, and hence,
results differ in bit rate, quality and latency. Results from encoders that use the same
compression standard may also vary because the designer of an encoder can choose to
implement different sets of tools defined by a standard. As long as the output of an encoder
conforms to a standard’s format and decoder, it is possible to make different implementations.
This is advantageous because different implementations have different goals and budget.
Professional non-real-time software encoders for mastering optical media should have the
option of being able to deliver better encoded video than a real-time hardware encoder for
video conferencing that is integrated in a hand-held device. A given standard, therefore,
cannot guarantee a given bit rate or quality. Furthermore, the performance of a standard
cannot be properly compared with other standards, or even other implementations of the same
standard, without first defining how it is implemented. A decoder, unlike an encoder, must
implement all the required parts of a standard in order to decode a compliant bit stream. This
is because a standard specifies exactly how a decompression algorithm should restore every
bit of a compressed video [3]>
4. H.264 LEVELS
The Group focused joint development in determining work H.264 to find a solution is simple
and flexible could include various applications through the use of a single standard, as is the
case in video standards, etc., and is this flexibility in the provision of facilities for several
profiles (represented groups of algorithms for the pressure data ) and levels (level private
suite of applications). The standard includes the following seven sets of capabilities, which
are referred to as profiles, targeting specific classes of applications
 Baseline profile (BP): It is primarily for applications with low cost resources with
physical entity and this file is used widely in applications of mobile devices and
applications to transfer data, such as video conversations.
 Main Profile (MP): This file is used for broadcast applications and storage, and the
importance of this file vanished when placing a high level to include those
applications.
 Extended Profile (XP): means the image streaming video, this file has the ability of a
relatively high pressure, and some additional possibilities to avoid data loss .
 High Profile (HiP): The primary file used for broadcast applications and disk storage,
particularly in the application of high-definition television and, for example,
applications in the HD DVD and Blu-ray .
  High 10 Profile (Hi10P): This file adds support for the previous file in decryption
process where they can processing 10 bits per sample of image resolution to decode
the data.
 High 4:2:2 Profile (Hi422P): used in professional applications that use interlaced
video, this file is used in the previous file basis (Hi10P) and added his coordination
shorthand chromatography with 4:2:2 sample while it uses 10 bits per sample per unit
decoding and gives the same quality.
 High 4:4:4 Predictive Profile (Hi444PP): used in the applications non are for the loss
of data was added to the previous file format support reduction chrome quality of
4:4:4 and support for samples up to 14 bits per sample, and also supports encoding
videos and pictures with tri-color separate.

5. TYPES OF F RAMES
H.264 consists of several different types of frames, such as ( I-P-B), and can be used for
encryption to get the required efficiency below illustrate the theoretical formula for each
quality of frames.
 (I-intra frame ) Is an autonomous framework which can encrypt and decrypt
independently without need for another picture as a source of information retrieval,
the first image of the video is for this type of frame, and the (I-frame) is the starting
point for the video display as well as his importance in information retrieval
synchronization if any damage in transport stream bit (bit stream), the flaw in this
window that consumes the largest possible number of bits for encryption because it
takes the window image full but on the other hand, the error rate is low. Encryption
method for this type of window has two properties, depending on the method of
dividing the cluster either ((16x16) or (4x4)) but in General is to convert the frame
version (RGB)format (YCbCr) and separated from the other components of the final
representation and is treated with a single image, so the representation of video
format ((4:2:0) YCbCr) is to reduce the sensitivity of the eye where the eye responds
to brightness by colors so the component (Y) represents the symbol of brightness
luminance while (CbCr) represents the color (chrominance) taken the element Y with
full size while the rest of the elements are reduced by deciding to half the amount of
action in the element size (Y) is (16x16) , the rest of the elements are the size of
(8x8), this means that embedded type of encryption key encryption process.
Encryption process as previously mentioned it is dependent on frame division,
divides the frame into multiple blocks of size (16x16) and has (4) types of encryption
as shown in Figures (eg, Figure 2).

Figure 2. The types of patterns to divide block (16x16)

 But the case of the split window to (4x4) , it has (9) types, as in Figures (eg, Figure 3)

Figure 3. Types of Patterns to divide block (4x4)

Choose a style for the adoption application and competence required for the
encryption process and the admissibility of the error rate, in most cases the amount of
the error rate of the video is of a higher flexibility compared to the error rate in the
case of a single image.
 (P-Inter Frame) Predictive Inter Frame: is derived from the current frame to the video
sequence frame by reducing the time between frames increase unlike previous quality
work only within the space of pixels, the principle of its work essentially compare the
block of the current window with the block of the previous frame and the centre of
block is search for match, this called (matching block), all theories have one and is the
best possible match and this is called motion estimation (ME), after finding the best
match, we put the block of the original block and the remaining known as compensation
(motion compensation), the link between the location of the current block with original
block is the transmission (motion vector (MV)) shown in Figures (eg, Figure 4).

Figure 4. The basic idea to represent the predictive inter frame (P-frame )

 B-frames (Bi-predictive inter frame), this type of frame be intermediate between (I,
and B frames) used at high levels for perfect efficiency but complex where the
highest of qualities as follows based on the comparison between more than one
source for block, meaning most forecasts from original source and source is
expected as in Figures (eg, Figure 5).
When retrieving the information in the decryption process is (I-frame), the former to
decrypt followed by (B-P frames) if used, the decryption depend one upon the other in
the information retrieval of the original frame. H264 has several ways in the encryption
that uses encryption (I-frame) or use (I&P ,(I&B&P)) and each method has its qualities,
if you use the first method, the quantity of bits encoded be high compared with other
cases but the error rate low because all the encrypted individually without relying on the
previous window, this method is used in some applications that need high resolution
cameras also in prisons and banks to get clearer picture during the up seizing process as
in Figures (eg, Figure 6) .

Figure 5. The representation frame (B)

Figure 6. Video encoding using spatial locations for pictures

But if you use the second method as shown in Figures (eg, Figure 7) they have
characteristics that they reduce the number of bits encoded and the error rate is
acceptable and this method is used in video compression in general and cameras , In the
third grade are more complex than both methods but with a reduced data encrypted and
contains a higher delay method because search matches more than one source.
 Figure 7. The first image is a frame ( Frame-I) and encodes a single, the second and third
picture only encrypts mobile part

6. Simulate Encryption and Decryption Using MATLAB

Previously we mentioned about the quality of the frames and how to process encoded in
H.264 encoding process is divided into two ways:

6.1 Encryption Process (I-frame)

I-frame is spatially encoded with a specific kind of styles in this research is the use of block
size (16 x16), divided into (8x8), Tables (eg, Table 1) shows real-time to encrypt and decrypt
the frame interconnection pattern within the first mode (Vertical mode) has been achieved
through Figures (eg, Figure 8). Each block is handled independently of and separately from
other 8x8 block are taken separately and applied the first style (vertical mode) and enters the
cosine (DCT). This conversion generates a representation of each block of (8x8) in the
frequency domain rather than the spatial domain. The resulting values of DCT process usually
consists of a few large values and many small values represent the relative sizes of these
transactions how important each block of information in the decryption stage of min after this
operation involved transactions to bring it shown in Figures (eg, Figure 8).

Table 1. Time encode and decode for I-frame

Video / 30 fps Time code /(sec/frame) Time decode (sec/frame)

Forman(288x352) 0.0383 0.034
Vipmen(160x120) 0.00052 0.00104
Bride(720x1280) 0.468 0.159
Piano(352x480) 0.054 0.04

Figure 8. The encrypted representation service

6.2 Encryption Process (P-frame)
We mentioned already that this type of encryption is encryption And this means that it uses
the correlation between the current frame and the frame (or frames) to achieve compression.
Tables (eg, Table 2) Shows real-time to encrypt and decrypt for P-frame, Temporal
encryption is achieved using vector motion (motion vectors.) the basic idea is to match each
block in the current window size (16x16) pixel in the frame of reference, the match here can
be calculated in many ways, but in H.264 uses a simple and more common is the sum of
absolute differences (SAD) offset from the current transaction to the last movement is
represented in reference to two values (horizontal, vertical vectors vector) are searched to find
the best tankers in the selected area to search only when a less value of the total difference
match between blocks this process is carried out under several algorithms one three-step
algorithm to search for blocks to neighbourhoods Match blocks beginning with predictive
search steps greater than half the search steps used in the previous methods and mechanism of
action as follows:
 Compared with nine points in each step, three score and six points on the sides.
 Under search by one space after each step and the search stops when the size of the
search space by one pixel.
 At each step a new search moving space research to find the best match for the
Center block from the previous search, blue circles of number one as in Figures
(eg, Figure 9) represent the first step of three steps when creating less differences
between the current and previous bloc begin the second step of the green circles
and also look on the differences is less up to the level of a single pixel of Orange
and last circles represent the end of the search for this theory.

 Figure 9. The theory of the three steps

Motion vector is a simple way to move a lot of information as shown in Figures (eg, Figure
10), but not always give an exact match to give the best quality, taking output subtraction
between the original frame and block the cluster framework forecast output encrypted as
shown in Figures (eg, Figure 11), the encryption process for residual image is similar to the
encryption (I-frame), but the difference is in the process of rounding and through practical
results as in Figures (eg, Figure 12) and found that the compression ratio is 70%of the original
size.
 Table 2. Time encode and decode for P-frame

Video Time code Time decode

30 fps (sec/frame) (sec/frame)
Forman(288x352) 0.0013 0.0011
Vipmen(160x120) 0.00032 0.001
Bride(720x1280) 0.0161 0.0054
Piano(352x480) 0.00186 0.00173

Figure 10. The motion vectors for Residual Image

Figure 11. The Residual Image

 Figure 12. The Decode Circuit

In Figures (eg, Figure 13) shows the encryption and decryption of video sequences
Forman and note that in the case of image encryption and decryption have same
properties.

Figure 13. The Original Foreman Sequence and Return Within Three-Step algorithm
7. CONCLUSIONS
DisplaysH.264 A major step forward in the field of video compression technology. and
provides techniques which enable better compression efficiency, due to more accurate
forecasting capabilities, as well as improving the ability to minimize errors. It provides new
possibilities for creating video encoders that managed to get high quality video and high
frame rates at per second and higher resolution at bitrates (compared to the preceding
criteria), and through the practical results of the MATLAB simulation was found that data
compression during real time up to(70%) of the video size Original part time implementation
261.89 s and time clk (4ns) and highest value can be obtained for a reference to the noise up
to (45 db) also shown in Figures (eg, Figure 14).

Figure 14. Simulink profile for H264/AVC

H.264 It is expected to replace other compression standards and methods used today, and
form became H.264 more widely available species in network cameras, video encoders and
video management software, designers of systems at the present time, the network video
products support both H.264 and Motion JPEG is perfect for maximum flexibility and
possibilities of integration.

REFERENCE
[1] A. Ben Atitallah, H. Loukil , and N. Masmoudi, FPGA DESIGN FOR H.264/AVC
ENCODER, International Journal of Computer Science, Engineering and Applications
(IJCSEA) Vol.1, No.5, October 2011, pp 119-138.
[2] Manjanaik.N, Dr.Manjunatha.R, Development of Efficient Intra Frame Coding in Advanced
Video Standard Using Horizontal Prediction Mode, International Journal of Emerging
Technology and Advanced Engineering, Volume 3, Issue 2, February 2013, pp 192-196.
[3] H.264 video compression standard, New possibilities within video surveillance, Axis
Communications, White paper, 2008.
[4] Amruta Kiran Kulkarni, Implementation of Fast Inter-Prediction Mode Decision in
H.264/AVC Video Encoder, Master Thesis, May 2012.
 Author
Mohammed H. AL-Jammas (Jun’02) born in
1966 in Mosul-Iraq. He awarded BSc in
Electronic and Communication Engineering
from the University of Mosul, Mosul-Iraq in
1988. Next, he awarded the MSc in
Communication from the University of Mosul,
Mosul-Iraq in 1994, and PhD in Computer
Engineering from the University of
Technology, Baghdad-Iraq in 2007. From
2002-2006, Dr. Mohammed worked with the
University of Technology in Baghdad. From
2007, he acts as an Assistance dean of the
College of Electronics Engineering at the
University of Mosul.
Through his academic life he published over 7
papers in field of computer engineering, and
information security.

Noor N. AL-Sawaf (August’08) born in 1988

in Mosul-Iraq. she awarded BSc in Electronics
Engineering from the University of Mosul,
Mosul-Iraq in 2010. Next, she awarded the
MSc in Electrical from the University of
Mosul, Mosul-Iraq in 2014.
Through his academic life he published 1
papers in field of image and video compress .

View publication stats