Você está na página 1de 9

Software Requirements

Specification
for

Facial Expression Recognition


System

Version 1.0 approved

Prepared by

Hafiz M.Abu bakar Siddique(BCSF06E96), Muhammad Jahangir


Javed(BCS-F07-M024),30

BS(CS) 8TH R

11/03/11

Copyright © 1999 by Karl E. Wiegers. Permission is granted to use, modify, and distribute this document.
Software Requirements Specification for <Project> Page ii

Table of Contents
1. Introduction................................................................................................................................1
1.1 Purpose ............................................................................................................................................... 1
1.2 Document Conventions....................................................................................................................... 1
1.3 Intended Audience and Reading Suggestions..................................................................................... 1
1.4 Product Scope..................................................................................................................................... 1
1.5 References........................................................................................................................................... 1
2. Overall Description....................................................................................................................2
2.1 Project Perspective.............................................................................................................................. 2
2.2 Project Functions................................................................................................................................. 2
2.3 Operating Environment....................................................................................................................... 4
2.4 Design and Implementation Constraints............................................................................................. 4
2.5 Assumptions and Dependencies......................................................................................................... 4
3. External Interface Requirements............................................................................................. 4
3.1 Hardware Interfaces............................................................................................................................ 4
3.2 Software Interfaces............................................................................................................................. 4
4. System Features......................................................................................................................... 4
4.1 Camera capturing................................................................................................................................ 5
4.2 Face recognition.................................................................................................................................. 5
4.3 Facial expression recognition............................................................................................................. 5
5. Other Nonfunctional Requirements.........................................................................................6
5.1 Performance Requirements................................................................................................................. 6
5.2 Safety Requirements........................................................................................................................... 6
5.3 Security Requirements........................................................................................................................ 6
5.4 Software Quality Attributes................................................................................................................ 6

Revision History
Name Date Reason For Changes Version
Software Requirements Specification for Facial Expression Recognition Page 1

1. Introduction

1.1 Purpose

The purpose of this project is to locate face into an image from video stream of webcam. Then
implement algorithms to detect the facial expression of person in that image. And finally respond
to that person in the form of some actions.

1.2 Document Conventions

1: font = times, size = 18, Decoration Bold, Style = Heading 1 Default setting of MS word for
main heading. This convention is used for Main Heading.
2: font = times, size = 14, Decoration Bold, Style = Heading 2 Default setting of MS word for sub
heading. This convention is used for Sub Heading inherited from each respected main heading.
3: font = Arial, size = 14, Decoration Italic. Used for Description of Sub Heading and related
explanation.
4: font = Arial, size = 14, Decoration Italic, Bold, Used to highlight the Specific description
features.

1.3 Intended Audience and Reading Suggestions

This document will be helpful to the different users and administrators and developer.

1.4 Product Scope

Facial expression recognition system is an important field in computer sciences. It is widely used
in robots, doctor helping systems, liar detection system, and advance integration in interrogation
system for intelligence agencies. As well as it used on airports to detect smugglers and terrorists.

1.5 References

1 IBM T. J. Watson Research Center, Hawthorne, NY 10532, USA. yltian@us.ibm.com


2 Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA. tk@cs.cmu.edu
3 Jagdish Lal Raheja Umesh KumarDepartment of Psychology, University of Pittsburgh,
Pittsburgh, PA 15260, USA.
jeffcohn@pitt.edu
4 Digital Systems Group, Central Electronics Engineering Research Institute (CEERI)/
Council of Scientific & Industrial Research (CSIR), Pilani-333031, Rajasthan, India 1
jagdishraheja@ceeri.ernet.in
umesh.kr1983@gmail.com
Software Requirements Specification for Facial Expression Recognition Page 2

2. Overall Description

2.1 Project Perspective

It is a new, self-contained product. Today terrorism is a big issue in the world. Most of the
countries facing most problems due to terrorism for example lost of life, damage of buildings,
increasing number of physiological patients and big problem related to security issues. So there is
a big need to make a system which can be help to recognize the terrorist in specific environment.
Pakistan has no such a system

2.2 Project Functions

This project has five major functions.


First it captures the image using the web cam.
Second function is facial image preprocessing, the face detection which can detect the human
face from the Captured image.
Third function is image normalization which is consists of four different functions for histogram
equalization, edge detection, thinning, and token generation.
Fourth function is facial feature extraction for making specific patter .These can be lies in
classification of serious, happy, thinking, sad, feeling pain moods.
The last function is the token matching and decision making called recognition module which
produces the final result.
Software Requirements Specification for Facial Expression Recognition Page 3

Flow chart of facial expression reorganization


1-Camera capture:
A Webcam can be use to capture an image of a person for analyses. Webcam should be a good
quality to improve performance.
2- Facial image pre-processing:
In the process of preprocessing, the first step is facial detection and localization. It means that
finds the position and face existence in the presence of background which mean that there can
be a complex detail or clear seen behind the person. The mean a big issue is difference between
face and background. So this problem must be eliminating.
3-Image normalization:
Image normalization is geometry normalization and gray normalization. The mean theme of gray
normalization is to improve image contrast.
4-Facial feature extraction:
After preprocessing, this system can make the feature extraction which can then further classify
into different categories for the purpose of recognition of facial expression. Correct recognition
rate of facial expression is directly depends on this part of overall system.
Software Requirements Specification for Facial Expression Recognition Page 4

5- Token matching and decision:


Finally match the generated data to the stored patterns data for final decision of specific person’s
mood.

2.3 Operating Environment

Pentium IV processor 2 GHz, 512 MB RAM and 1.3 mega pixel camera is required for this
system.

2.4 Design and Implementation Constraints

This system is developed in visual C++ using libraries of open CV. Microsoft window 7 is primary
operating system used for develop this system.

2.5 Assumptions and Dependencies

1- Capture the image from front side of the person.


2- Capture the image in brighten environment. If capture the image in night mode then face
in the image should be clear.
3- Capture the image of the single specific person at the spot.

3. External Interface Requirements

3.1 Hardware Interfaces

1-webcam: used for capturing image of a specific person.


2-Computer: To process and analyses of captured image for required results.

3.2 Software Interfaces

Open CV 2.2 libraries will be use for image processing, which is developed by Intel to enhance
the capabilities of graphic hardware of computer system. It is available in market to use on the
bases of GPL license for free, commercial and non-commercial use. Microsoft Windows 7 will be
use as operating system for development of this system. Visual C++will be use as development
language and C# will be use as interface design. MS SQL 2005 will be use as database to store
the facial expressions patterns of a specific person. There may be more than one person’s
related data but there only one person data will be use for initial testing.

4. System Features
The major services provided by the product are following.
Software Requirements Specification for Facial Expression Recognition Page 5

4.1 Camera capturing

4.1.1 Description and Priority


System will capture the image from video stream of webcam. Captured image
should be high quality for accurate result Camera image capturing is high priority (1)
because analyses will be real time and result will be on the spot. Webcam can be
integrated into a notebook or a separate having cost approximately 1200 Rupees.
4.1.2 Stimulus/Response Sequences
First user launches the application. Application automatically detects the best
possible camera attached with computer, and begins capturing the images from
video stream of camera. If user wants to view video stream on the screen it is
optional.
4.1.3 Functional Requirements
REQ-1: Camera should be attached with computer to continue the process if
camera is not attached with computer it will give alert for absence of camera.
REQ-2: Secondly environment should be clear for clear vision. If there is some
problem related to clear camera image capturing, it then also give alert for it.
REQ-3: Capture the image in the presence of light so that face of the person should
be clear.
ReQ-4: Person should be in front of camera and face is directly to the camera.

4.2 Face recognition

4.2.1 Description and Priority

System will recognize and locate the face in captured image. Locating the position
and face existence in the presence of complex or simple background which mean
that there can be a complex detail or clear seen behind the person. The priority of
this feature is medium (2).

4.2.2 Functional Requirements

REQ-1: Image of the person must be captured.

4.3 Facial expression recognition

4.3.1 Description and Priority

System will recognize the facial expression of person from given person’s image by
applying algorithm. It recognizes whether the person is happy, sad or fear etc.

4.3.2 Functional Requirements

REQ-1: Person face should be recognize from captured image.


Software Requirements Specification for Facial Expression Recognition Page 6

5. Other Nonfunctional Requirements

5.1 Performance Requirements

Computer should be Pentium IV or above for best performance of software. System should be
virus free. Make possible any heavy application is not running on the computer during running
this software. Computer RAM is main factor of software performance so RAM should be
maximum according to software requirements.

5.2 Safety Requirements

Camera that is use for this software should be plug-in properly. If external camera connected to
computer then connector and table must not damage. If camera not needs to use further, turn of
the camera, if using external camera then unplug form computer properly to safety from lose that
can be cable damage or some circuit problem. If using infrared camera then should not use it for
long time. If not needs to use software further then properly close the software for the safety of
environment which cause the increasing of environment temperature.

5.3 Security Requirements

Software is password protected. It will store historical data related to specific person so only
authorize user should be access this software. If not care about privacy historical data may be
steel. Database contain software related data so do not try to open the database directly which
cause lost of data.

5.4 Software Quality Attributes

1-Reliability:
Software is reliable if train it maximum by providing large number of patterns of facial expression,
which helpful in facial expression recognition. Good quality camera is best for reliable and
accurate result.
2-Portibility:
Software is portable. It can be use at any computer and easily move from one computer to
another computer using USB Flash.
3-Avialibility:
Software is for a single system use at a time. It is cannot use on network like LAN and internet.
Software will use only one camera at a time. If more than one cameras are attached to system it
will use best possible camera.

Appendix: Glossary
USB: Universal Serial Bus.
LAN: Local Area Network.
GPL: General Public License
Software Requirements Specification for Facial Expression Recognition Page 7

RAM: Random Access Memory.

Você também pode gostar