Facial Recognition application in computer system

Transcription

1

2 Faculty of Information Technology Department of Computer Science Second Semester Project Progress Report 2013/2014 Facial Recognition application in computer system Birzeit University Prepared by: Karima Kook Sana Sheikh Supervisor: Mr. Ibrahim Sarahin 2014

3 Abstract Human face detection and recognition by computer systems has become a major field of interest. Face detection algorithms are used in a wide range of applications, such as security control, video retrieving, biometric signal processing, human computer interface, face recognitions and image database management. However, it is difficult to develop a complete robust face detector due to various light conditions, face sizes, face orientations, background and skin colors. In this report, we propose a face detection method for color images. Our method detects skin regions over the entire image, and then generates face candidates based on a connected component analysis. Two of the most important aspects in the general research framework of face recognition by computer are addressed here: face and facial feature detection, and face recognition or rather face comparison. The best reported results of the mug-shot face recognition problem are obtained with elastic matching using jets. In this approach, the overall face detection, facial feature localization, and face comparison is carried out in a single step. This report describes our research progress towards a different approach for face recognition. On the one hand, we describe a visual learning technique and its application to face detection in complex background, and accurate facial feature detection. On the other hand, a fast algorithm for 2D-template matching is presented as well as its application to face recognition. Finally, we report an automatic, realtime face recognition system. II

4 Table of Contents Abstract.. ii Acknowledgment..... vi Chapter one : introduction Aims and objectives Aims objectives motivations overview of the technical area outline.6 Chapter two : History History Details of relevant theory Review Of The Past Reports US-VISIT Faces of Facebook Detection and Recognition Face Recognition And Detection Brief introduction of the proposed and the solution Correlation Some Application Of The System Advantages and Disadvantage Of The System advantage of the system disadvantage of the system. 14 Chapter three : System analysis, system model, and system architecture Requirements functional requirements III

5 3.1.2 non functional requirement s entity relationship scenario log in use case for an exist user sign up use case for an exist use sign up use case for an exist user Use Case Diagrams log in use case for an exist user, sign up use case for an new user sign up use case for an admin Use case name: Face detection Sequence Diagram Actors Activity Diagram Activity Diagram for registration (based on registration use case) Activity for taking picture: Activity Diagram for log in use case Class Diagram ( in general ) Data Flow Model Chapter 4 discussion Discussion what we should do in next semester Chapter 5 conclusion..33 Glossary Reference IV

6 List of Figures Figure corelation..12 Figure log in case Figure sign up case 20 Figure admin case Figure use case name face recognition 21 Figure sequence diagram...23 Figure activity diagram for registration. 25 Figure activity diagram for taking picture Figure activity diagram for log in use case Figure class diagram V

7 Acknowledgment We would like to express our deeps thankful and gratitude to our supervisor MR. Ibrahim Sarahin who helped us in each step in this report, most sincerity to his support and guidance throughout the research. Special thanks for our teachers that without them we can t complete this project. We would also like to extend our deepest gratitude to families and friends, without their encouragement, we would not have a chance to be at Birzeit University. VI

8 Chapter 1 Introduction Section 1: Aims And Objective Section 2: Motivation Section 3: Overview Of The Technical Area Section 4: Outline - 1 -

9 I. Introduction In our recent years, with the huge development, and with these grate technological life, security become very important issue in each part of our life, as a result of that there are several way to protect our properties, like password and user name to each member and so on Here, we will explain one of protect way that let you enter into your owns thing or your porous which depend in special algorithm, it is easy, quickly, accurately and guaranteed. it is called Facial Recognition way Face recognition is a task so common to humans, that the individual does not even notice the extensive number of times it is performed every day. Although research in automated face recognition has been conducted since the 1960 s, it has only recently caught the attention of the scientific community. Many face analysis and face modeling techniques have progressed significantly in the last decade. However, the reliability of face recognition schemes still poses a great challenge to the scientific community.[1] - 2 -

10 1.1 Aims And Objective Aims : The general aim in face recognition application that implement our project at the best way to be useful by using our knowledge in many courses, using general model like waterfall model which lead to solve many problems step by step. to provides a better representation and achieves lower error rates in face recognition Objective : It is have to be more specific term, to focus about the main idea and the details to applicant facial recognition in our project, that easier to all users, most quickly, and modern application. The objectives of this thesis will be: To discuss and summarize the process of facial recognition. To look at currently available facial recognition techniques. To design and an UML analysis of this system and implement it depend in our assumption. To implicate what we learned from software course to present complete analysis business in UML diagrams. To design and develop a robust facial recognition algorithm, so the algorithm should be usable in a simple and easily adaptable setup. This implies a single camera setup, preferably a webcam, and no use of specialized equipment

11 1.2 Motivations Recently, Face recognition has received a big attention and huge interest from the scientific community as well as from the general public. The interest from the general public is mostly due to the recent events of many events around the world, which has increased the demand for useful security systems in each term of their life, so Facial recognition applications are far from limited to security systems. To construct these different applications, precise and strong automated facial recognition methods and techniques are needed. However, these techniques and methods are currently not available or only available in highly complex, expensive setups. The topic of this thesis is to help solving the difficult task of robust face recognition in a simple setup. Such a solution would be of great scientific importance and would be useful to the public in general

12 1.3 Overview Of The Technical Area The technical details include many aspects, this application is a desktop application that have an interface which is receive the data and represent the connection between the users and the system, it is have database files the system depend on programming languages so we will use IDE Java (net beans) and software tools especially the UML from Wikipedia UML tool or UML modeling tool is a software application that supports some or all of the notation and semantics associated with the Unified Modeling Language (UML), which is the industry standard general purpose modeling language for software engineering. [3] Which is the most important issue in any application that help any designer in his project

13 1.4 Outline I naturally divided into five parts, where each part depend in the previous part and requires knowledge from the preceding parts. Chapter one: In chapter one we presented a brief introduction, then we talked about goals, aims and motivation in this report. Chapter two: Chapter tow includes an overview of the technical issues that we will used it to implement the system that we are working on. That desktop application with database using programming language and software issues. Chapter three: Development, includes the details, summary of past reports and books, then summary about the proposed work, the choices, the solutions. Chapter four : This chapter include the discussion of the report depend in our research and opinion, with summary of work done, the critical point in general. Chapter Five: Finally, in chapter five we represented a conclusion n and the next work that will be done in next semester

14 Chapter 2 Face Recognition Application In History (background) Section 1: Review Of The Past Reports Section 2: Face Recognition And Detection Section 3: Brief introduction of the proposed and the solution Section 4: Applications Section 5: Advantage and Disadvantage - 7 -

15 2.1 History Face recognition systems have been conducted now for almost 50 years, face recognition is one of the researches in area pattern recognition & computer vision due to its numerous practical applications in the area of biometrics, Information security, access control, law enforcement, smart cards and surveillance system. The first large scale application of face recognition was carried out in Florida Details of relevant theory 1- US visit US-VISIT s biometric identification and analysis services are revolutionizing the way decision makers establish and verify a person s identity. Biometrics unique physical characteristics, such as fingerprints, that can be used for automated recognition form the foundation of US-Visit s identification services because they are reliable, convenient and virtually impossible to forge.[4] 2- Faces of Facebook: Identified profiles in online social networks Individuals using their real first and last names on Facebook, LinkedIn, Google+, etc. Continuing improvements in face recognition accuracy

16 2.2 Review Of The Past Reports US-VISIT "US-VISIT", and in order to prevent the entry of some of the people suspected of having links to terrorist organizations. Face Recognition System is an application to verify a person's identity by comparing pictures that has been taken recently from a digital camera or VCR, and compare it with images stored in the system database. And this application used widely in recent times for the purposes of security and access control, like other existing security systems that rely on biometrics, such as fingerprints or iris of the eye. Where it is in the United States was using the program to get rid of fraud and swindling, arrest suspects,but he recently used this program face recognition system in the granting of visas and delivery. When the traveler receives a visa are fingerprinted as well as his personal image, is verified by the owner of the fingerprint and the image that he is not one of the suspects by comparing his image and mark a database of known criminals including questionable. When the traveler arrives in the United States at the port of entry, is to use those same pictures and fingerprints to verify that the person who received the visa is the same as the next person to give permission before entering. As well as a multitude of uses in banks, airports : In the bank uses facial recognition machines to withdraw money, with software to verify the client before you bring the money to him, and could be dispensed card to withdraw money and sufficiency of standing in front - 9 -

17 of the machine, which will recognize you and you open your account for you drag or deposit. ATM. So at the airports it is currently experience in this system registered traveler program called Registered Traveler, where the program allows for a traveler to accelerate the inspection procedures and pass through security barriers customize special tracks these people. Enrolled in this program, where they are identified by special cameras installed on these routes by face recognition systems. In companies the using of face recognition system to controlling the presence and departure of employees in companies and institutions have developed a huge company of A4Vision integrated system for this purpose : Faces of Facebook Faces of Facebook is the most general application in face recognition so Hence, face recognition creates the potential for your face in the street to be linked to your online identity, as well as the sensitive interfaces that can be made about you after blending together offline and online data. Faces of Facebook research focus on some concept: recognize the feasibility of combining publicly available online social network data with off the shelf face recognition technology for the purpose of large scale, automated, and this is based on two things: individual re identification, online and offline and " accretion " and linkage of online, potentially sensitive, data to to someone s face in the offline world

18 2.3 Detection and Recognition Detection: that distinguish the faces in any image that is received. Recognition: to identify that this is that intentional face we want to study it, and for who this face, there is any name of our safely file can have this face, which is the highest percentage of matching and for who? The three primary face recognition tasks are: 1. Verification (authentication) - Am I who I say I am? (one to one search) 2. Identification (recognition) - Who am I? (one to many search). 3. Watch list - Are you looking for me? (one to few search) Face Recognition And Detection A face recognition system based upon the previously described face and facial feature detection/tracking system. The recognition program, which runs a module of the face tracking system, uses template correlation to compare detected and tracked faces from the video sequence with the database. Every time a person is successfully detected and tracked, the recognition module is launched. If the confidence level of the recognition is not higher than some threshold, a new entry is added to the database. Entries are later associate with the individual information such as the name. A database of templates is constantly used and updated by the system. The templates are created using the outer eye corners as the

19 reference to normalize the scale and the rotation of the input face. Note that more than one entry per person is allowed in the system, and they are automatically created if the person changes his or her appearance significantly with respect to previous entries. This way, in a very simple scheme, the system automatically learns from video and the individual changes due to facial expressions.[3] 2.4 Brief introduction of the proposed and the solution The underlying idea behind these approaches is to reduce a facial image containing thousands of pixels before making comparisons. To do this, a face image is transformed into a space that is spanned by basis image functions, just like a Fourier transform projects an image onto basis images of the fundamental frequencies Correlation Two images are superimposed and the correlation between corresponding pixels is computed for different alignments. And this figure example of correlation Fig

20 2.5 Some Application Of The System 1- Entertainment/ Human-Computer Interaction: Video Gamingvirtual reality- training programs. 2- Smart Card applications : (in lieu of maintaining a database of facial images, the face-print can be stored in a smart card, bar code or magnetic stripe, authentication of which is performed by matching the live image and the stored template) 3- Security : access control to buildings, airports, ATM machines, etc 4- law enforcement and Surveillance : Advanced video surveillancecrime stopping and suspect alert-suspect tracking and investigation- suspect background check - post event analysis shoplifter recognition

21 2.6 Advantages and Disadvantage Of The System Advantage of the system : 1- Face recognition systems are the least intrusive from a biometric sampling point of view, requiring no contact, nor even the awareness of the subject. 2- Photos of faces are widely used in passports and driver s licenses where the possession authentication protocol is augmented with a photo for manual inspection purposes; there is wide public acceptance for this biometric identifier. 3- The biometric works, or at least works in theory, with legacy photograph data-bases, videotape, or other image sources. 4- Face recognition can, at least in theory, be used for screening of unwanted individuals in a crowd, in real time. 5- It is a fairly good biometric identifier for small-scale verification applications Some Disadvantage of the system : 1- face recognition system can't tell the different between identical twins. 2- problem with false rejection when people change their hairstyle, grow or shave a beard or wear glasses

22 Chapter 3 System analysis, system model, and system architecture Section 1: Requirements Section 2: Entity Relationship Section 3: Data - Flow Model

23 3.1 Requirements Requirements analysis is critical to the success of a systems. Requirements include descriptions of system properties, specifications for how the system should work, and constraints placed upon the development process. Requirements come from customers and sometimes from developers. Customers, are the people who provide a description of what they want from the system do and are paying for the development of the system. Developers might have requirements related to system performance and other technical topics. So there contract between the customer and the software developers on the product that will be delivered, called a requirement document. Requirements are divided into two basic types are functional and non-functional requirements Functional Requirements Functional requirements define what the system or application will do. For example, the system must send a an whenever a certain condition is met. Functional requirements may be calculations, technical details, data manipulation and processing and other specific functionality that define what a system is supposed to accomplish. In our system's Functional Requirements include the following: 1. Acquiring face images The photos should be taken with a computer camera. many photos should be taken for each user. The software should make the user look in different

24 directions. The majority of photos should be taken for a neutral expression face while some for smiling and annoyed expression face. There should be some flags which the user would be instructed to look at when taking a photo. 2. Pre-processing image Given a face image, the software should remove background, clip and scale the image to a certain size of the pixels with the face centered. 3. Detecting features of the face Given a pre-processed face image, the software should apply a center-surround-filter to the image; the output image would be the face with its features (eyes, nose, mouth, and so on) outstanding. Noise removal correcting brightness 4. Managing users face images 5. Finding useful features for recognition Non-Functional Requirements Non-functional requirements place constraints on how the system will work, and is a requirement that specifies criteria that can be used to judge the operation of a system, and describe system attributes such as security, usability, reliability, efficiency, privacy and accessibility. In our system's Non Functional Requirements include the following: 1. It should be extremely convenient to take a photo

25 For example, once the user clicks on Taking a photo button, the snapshot would be done and the image file would be saved immediately. It should just need a one-click effort. 2. Instructions for acquiring face images should make users clear enough as to where they should look at when taking photos. 3. Usability, scalability, efficiently and security. 3.2 Entity Relationships Scenarios Scenarios describe how users of the system interact with the system to achieve their particular tasks, usually refer to representative instances of user-system interactions log in use case for an exist user Karima want to login to her laptop, the system will ask her to take a photo of her face by laptop s camera or she can enter hardware photo. The computer system will send the facial photo to the database and compare it with the store one, if it match, then,the file will be shown, if not the system will let her retake the picture or get it out of the system

26 sign up use case for an exist user Sana' want to Sign up from her laptop, the system will ask her to take a photo of her face by laptop s camera to add the photo to registration setting. Here the computer will send request to make new account and let her register, then new account will added to the users list in the database sign up use case for an exist user The admin is responsible for the management of the system, writing the algorithm about the system, make controlling and updating the version Use Case Diagrams : log in case for an exist user : Fig

27 log in case for an new user : Admin Case : Fig Fig

28 Use case name: Face detection: The use case of face recognition below explain the authorities of exist user and new user, that new user the only function can do it to sign up to create new account and add it to the list as a photo, in the other hand the exit user can do many functions that allow him to log in to the system, take photo by compute s camera or enter a facial picture, then the system will make comparison between the stored photo and the photo that entered by exist user it in the system,if the correlation okay the exist user can here do many function as delete, add, edit, encryption or etc,files. Fig

29 3.2.3 Sequence Diagram The figure below explain the sequence diagram that the objects is: Interface, desktop application, face recognition system, Data base file. The new user will make new registration as a sign up by interface, the registration requirements will send to the desktop that will send it to the facial system and store it in the data base. Now new account will add to the list in the data base. The exist user will make sign in because already has an account, he will take a photo by interface which has a computer s camera that enable the user to take a picture if the camera not work the user can enter any facial picture in some constraints then the photo will be sent to the facial recognition system and it is task to matching the photo inside facial recognition system object this enable the system to make comparison between the entered photo and the stored one in the data base, if the matching were more than 70% the system will allow the user to see the file by desktop, if the matching were less than 70% the system will not allow the user to see the file, of the system will allow this user see the files but the files will show as encryption

30 Fig

31 3.2.4 Actors: The actor in Unified Modeling Language who specifies the roles in each case played by a user or any other system that interact with subject. So the actor could be several depend in many cases: 1- New user : this actor can only make new registration to the system. The actors will be (interface, desktop, face recognition System, database) 2- Exit User : this actor enable to login to the system and see the files 3- The object will be: ( interface: that allow any user to see the general screen, Desktop Application : will be the middle step between interface and the face recognition System that will add it to the new database ) 4- Admin : the only one who can be more controller of the system we can say that is the root. The actors will be (interface, desktop, face recognition System, database, and control)

32 3.2.5 Activity Diagram Activity Diagram for registration (based on registration use case) Activity for new user In this diagram, if new user wants to use system, first should submit your name, and then system register him by storing name and image in database in this way the user has an account for using system again. Fig

33 Activity for taking picture: This activity diagram taking image for each user either existing or new user, the system editing photos from background and any other color. After that make sure that images matching with users faces,if matching store in database. Fig

34 Activity Diagram for log in use case The activity diagram for login user use case, the client already has an account,just enter name and system matching this information with his/her face, if matching successfully,the user logging system and can making any changes, otherwise, the system ask user to logging system gain. Fig

35 3.2.6 Class Diagram ( in general ) Fig Image Manipulation Class: This class acts as an interface of this domain, all the inter-domain data communication must be through this class. Pre-processing Class: This class pre-processes the images, including removes the background, scales the size of the image, and converts image to grey scale. the Image-Shop Class: This class provide a set of useful methods for pre- processing images

36 Face Class: This class detects the boundary of the face in a image and counts the number of pixels in the image. Settings Class: This class stores all the settings information that can be changed during the runtime of the software. This information includes the standard size of a pre-processed image, the number of pixels of an image, the settings for background removal, and the default output directory and so on. Managing Class: This class manages users ID numbers. Acquiring Class: This class acquires images and saves the acquired image to database. 3.3 Data-Flow Model Main flow (Basic flow) for sign in use case 1- User Authenticates 2- User select to recognize an image 3- FRS (Face Recognition System prompts user to select RFS face. 4- Data record of profile containing Trained Face which has highest recognition confidence is displayed

37 Pre-conditions 1- The client must be connected to the server 2- The specified profile must exist Alternative flows (if any) It s not exist Exception flows (if any) 1- Input is not valid 2- Not able to authenticate 3- Request a profile that you are not authenticated for ( denied access to the profile ). Special Requirements (if any) Post-conditions (Exit Conditions) Cleaver camera Easy equipment User login system successfully and make changes on files Data record of requested profile is successfully updated

38 Chapter 4 Discussion

39 4.1 Discussion The facial recognition application appealed to the public group for help identifying them. Why we need it? This new system will improve measures to protect people s identities and help detect identity fraud. It will also improve road safety by supporting the one person one license principle, which ensures drivers are made accountable for their driving behavior 4.2 What we should do in the next semester: Present an algorithm that capture the faces from the pictures, make definition for this face with the name of the person in the data base, e.g. of that how the system of the Facebook can recognize the faces and make tags for their owns automatically it is just need t defended before, from any one. so use it as a imprint for entering to the computer, create accounts for customers in bank, entering by credit card. It is just a very important system with high level of security more than the password, in computer multiple users can use the same desktop but the files where be encrypted unless if their own com and open them in this way. The problems we faced and that we will try to solve : The changing in the environment that was the picture has be taken will causes some problems in each time that the user will take a picture to enter like the color of skin a, the face situation in many cases. -Taking the picture quickly -The Facelift

40 Chapter 5 Conclusion

41 5. 1 Conclusion We have presented the definition, explanation and more details about face detection and recognition in any image or any user want to login to the computer so face images depend on the color mainly that uses color details, connected component analysis. Our method uses compensates for the Lighting condition of the image, and overcomes the difficulty of separating faces that are connected together using image morphology processing. Finally we presented an UML analysis that explain the objects and shows every aspect in our report that will to implementation in future work

42 Glossary In this article we will have a keyword Keywords. Face detection, face recognition, facial feature detection, Maximum likelihood, maximum discrimination, information theory, visual Learning techniques, template matching

43 Reference [1] march, 27,2014 [2] : Face Recognition Algorithms, Proyecto Fin de Carrera [3]: [4]: us visit,enhancing Security Through Biometric Identification [5]: Faces of Facebook: Privacy in the Age of Augmented Reality, Alessandro Acquisti, Ralph Gross, Fred Stutzman [6] : April, 16, 2014 [7] : Finding face by motion, Robert W.Frischholz [8]: Face Detection and Recognition, Antonio J. Colmenarez and Thomas S. Huang. [9] : Face Recognition Methods & Applications,Divyarajsinh N. Parmar1, Brijesh B. Mehta2 [10] : A Survey of Face Recognition Techniques, Rabia Jafri* and Hamid R. Arabnia* [11]: Face Detection Project Report, Ana Bertran, Huanzhou Yu, Paolo Sacchetto. [12]: How Facial Recognition Systems Work, by Kevin Bonsor and Ryan Johnson