Docstoc

Biometric Attendance System---IEEE2011

Document Sample
Biometric Attendance System---IEEE2011 Powered By Docstoc
					           Proceedings of the 20 1 1 IEEEIICME
International Conference on Complex Medical Engineering
                May 22 - 25,Harbin,China


                                           Biometric Attendance System
                                              Engr. Imran Anwar Ujan and Dr. Imdad Ali Ismaili
                                                 Institute of Information & Communication Technology,
                                                      University ofSindh, Jamshoro, Sindh, Pakistan
                                                                         Pakistan


      Abstract - This research work has application for attendance                         junction. In this research, we will be dealing mainly with ridge
 system of employer's and students in general. The system will                             endings and bifurcations.
 facilitate institutions! organization to make attendance individual                             There      are   various    types      of   approaches      proposed    in
 in time along with data information thumb impression will be
                                                                                           literature       for   both   image       enhancement,         and     minutiae
 taken as a signature for the system entry. Main design and
                                                                                           extraction from fingerprints. The literature on these techniques
 challenge in this system is the design of database architecture and
                                                                                           will be examined are reviewed in determining the best
 its business logic.
                                                                                           approach to develop for this research. In particular, the
                          I. AIMS AND OBJECTIVE                                            fingerprint image enhancement algorithm employed by Hong
                                                                                           et al. will be evaluated and implemented to understand how
       The aim of this system is to implement in C#.net set of                             the enhancement algorithm works and how well it performs.
 reliable techniques for fingerprint image enhancement and                                 Once      a reliable      minutiae     extraction     technique      has   been
 minutiae extraction. The performance of these techniques will                             implemented and tested, this can be used as the basis of
 be evaluated on a fingerprint data set.                                                   statistical analysis of fingerprint minutiae.
 In combination with these development techniques, statistical                                   The work of Tu and Hartley and Pankanti et al. can be
 experiments can then be performed on the fingerprint data set.                            examined in which a statistical framework for analyzing
 The results from these experiments can be used to help us                                 system performance has been presented. Tu and Hartley
 better   understand       what     is    involved    in   determining         the         defined a means of forming a binary code from set of
 statistical uniqueness of fingerprint minutiae.                                           fingerprint features and then performing a set of matching
    The     main   aim     that   this      system    would     test     whether           experiments on the database to estimate the number of degrees
 attendance by fingerprint is enough for identification. It is                             of freedom within the fingerprint popUlation.
 expected that the work in this system will reach the stage of
 being able to fully test hypothesis.
                                                                                           Problem
                    II.    BACKGROUND/CONTEXT                                              There      are     some    problems          which   face    by      fingerprint
                                                                                           recognition or identification security. We can catch a cold by
  Fingerprints are the oldest form of biometric identification.
                                                                                           touching a biometric system (fingerprint).
 Modem fingerprint based identification is used in forensic
                                                                                                 •    Twins       have      identical    biometric     traits    (identical
 science,    and    in     biometric        systems     such        as   civilian
                                                                                                        fingerprints, irises ... ). This is the same clones.
 identification    devices.       Despite      the    widespread         use   of
 fingerprints, there is little statistical theory on the uniqueness                              •    Stolen body parts can be reused.
 of fingerprint minutiae.                                                                        •    Biometric features can be reconstructed from the
       A fingerprint is formed from an impression on a surface of                                       template.
 composite curve segments. A ridge is defined as a single
                                                                                                 •    Making a fake finger is easy.
 curved segment, and a valley is the region between two
 adjacent     ridges.     The     minutiae,      which        are    the   local
                                                                                                The inability of fingerprint systems to enroll children and
 discontinuities in the ridge flow pattern, provide the details of
                                                                                           small Asian women.
 the ridge-valley structures, like ridge endings and bifurcations.
 There are 50 to 150 minutiae on a single fingerprint image.                                                      III. PROJECT SPECIFICATION
 Features such as the type, direction, and location of minutiae
                                                                                           A.    Project Description
 are taken into account when performing minutiae extraction.
                                                                                                 The goal of this project is to daily attendance of employee
       The work of F.Galton defined a set of Galton Features for
                                                                                           through fingerprint. The project is design and implements
 fingerprint identification, which since then, has been refines
                                                                                           software architecture for fingerprint analysis. The system
 and    extended    to    include        additional   types    of    fingerprint
                                                                                           should be able to extract key features                      from a scanned
 features. However, most of these features are not used in
                                                                                           fingerprint image and to compare these with a database of
 automatic fingerprint identification systems. Instead the set of
                                                                                           known fingerprint images and/or extracted feature sets.
 minutiae types are restricted into only two types, ridge endings
                                                                                                 For this project we provided with a set of previously
 and bifurcations, as other types of minutiae can be expressed
                                                                                           acquired fingerprints and a working fingerprint sensor with
 in terms of these two features types. Ridge endings are the
                                                                                           driver software for Windows. Our expectation had fulfilled by
 points where the ridge curves terminates, and bifurcations are
                                                                                           most of the algorithm development which executed in C# dot
 where a ridge splits from a single path to two paths at a Y-
                                                                                           net and this work done on a Windows PC.




  978-1-4244-9324-1/11/$26.00 ©2011 IEEE                                             499
B.    Project Task                                                                           of project parameters are tentative, the plan will always need
     The project can be split into a set of principle tasks                                  to be modified.
representing a progression towards the end goal of a working                                       A    structure         of    BAS      software        development              plan      is
fingerprint analysis system.                                                                 described below
      1.   We had must reviewed techniques for analyzing
                                                                                                                               IV. INTRODUCTION
           fingerprints and performing pattern recognition on
           sets of fingerprints. Several of the most promising                                     Here we describe the brief objectives of the BAS project
                                                                                             and       set    out       the    constraints      which      affect          the         project
           algorithms/ techniques had implemented in C# .net
                                                                                             management.
           and initial testing performed on the test set of
                                                                                             A.    Objectives
           fingerprint image provided.                                                              •         Computerize the daily attendance system
      2.   The        Biometric       Attendance         System             software
                                                                                                    •         Attendance by fingerprint
           architecture for the main system was designed; the
                                                                                                    •         Protect         the   proxy       which     IS        doing        III    daily
           main subsystems required were determined and a
                                                                                                              attendance
           method of implementing a full system was evaluated;
           work       architecture    and   several          of    the    functional         B.    Constraints
           subsystems were implemented.
                                                                                                   •         Budget

      3.   We     analyzed     algorithms       were         implemented        and                                1.    Veridicom Fingerprint Sensor RS.8,500

           integrated with the fingerprint sensor, and real-time                                                   2.    Expenditure         for   collection         of     data         and

           acquisition       and     analysis    of      a        fingerprint   was                                      information Rs.2500

           demonstrated; an improvements in processing speed                                                       3.    Total budget is Rs. 1 1 ,000

           and implemented and demonstrated.
                                                                                             C.    Methods
      4.   Improvements in the analysis of an acquired image
                                                                                             A rough task breakdown for this project is as follows:
           may        be   achieved      through         image           processing;               •         Examine and review available literature on image
           combining multiple acquired images to provide an
                                                                                                             enhancement and minutiae extraction techniques.
           enhanced composite image; or more sophisticated
                                                                                                   •         Develop a series of image enhancement techniques to
           statistical or mathematical approaches.
                                                                                                             aid the minutiae extraction process.
      5.   Improvements in pattern matching may be achieved
                                                                                                   •         Develop a set of reliable techniques to extract the
           through various pattern recognition approaches; the
                                                                                                             minutiae from fingerprint images.
           students        should     evaluate        several            approaches,
                                                                                                   •         Evaluate the performance of the techniques using the
           developing an evaluation methodology which enables
                                                                                                             fingerprint data set.
           a comparison in terms of improved recognition and a
                                                                                                   •         Use    existing        techniques      as    the       benchmark              for
           reduction in terms of false positives and negatives.
                                                                                                             comparing          the    performance             of    the     technique

Integrating the techniques of 4 and 5 with the real-time                                                     developed.
acquisition of fingerprints will add significant bonus value.                                      •         After reliable minutiae detection techniques have
                                                                                                             been developed and tested, then statistical analysis
C.    Project Planning                                                                                       experiments on            the fingerprint data                set     can     be
      Effective management of a software project depends on
                                                                                                             performed and documented.
thoroughly       planning     the     progress      of       the     project.   We
anticipated problems which arose and prepared solutions to
                                                                                                                                (Employee Fingerprint Table)
the project problems. A plane, drawn up of a project, we used
                                                                                                                                 EmpfPrlnt
as the driver for the project. The initial plane evolves as the                                                                 [j]EWlD
project progress and better information.                                                                                        O FPRlNT
      The planning process starts with an assessment of the
constraints      (required    delivery      date,     overall        budget,    etc)
                                                                                                       (Employee Profile Table)                          (Attendance Table)
affecting the project. This is carried out in conjunction with an
estimation of project parameters such as its structure, size, and
distribution     of     functions.    The    progress             milestones    and
deliverables are then defined. The process then enters a loop.
A schedule for the project is drawn up and the activities
defined in the schedule are initiated or given permission to                                                                            Fig.1
continue. After some time usually about 2-3 weeks, progress
is reviewed and discrepancies noted. Because initial estimates




                                                                                       500
                                                                                            orientation and ridge frequency parameters. However, in
                                                                                            practice,   this     does   not     pose   a   significant    limitation   as
                                                                                            fingerprint     matching          techniques    generally     place    more
                                                                                            emphasis on the well-defined regions, and will disregard and
                                                                                            image if it is severely corrupted.

                                                             (Fail)
                                                                                                  Overall, the results have shown that the implemented
                                                                                            enhancement algorithm is a useful step to employ prior to
                                                                                            minutiae extraction. The Crossing Number method was then
                                                                                            implemented to perform extraction of minutiae. Experiments
                                                                                            conducted have shown that this method is able to accurately
                                                                                            detect all valid regions, and will disregard an image if it is
                                                                                            severely corrupted.
                                                                                                   However, there are cases where the extracted minutiae do
                                                                                            not correspond to true minutiae. Hence, an image post
                                                                                            processing stage is implemented to validate the minutiae. The
                                                                                            experimental results from the minutiae validation algorithm
                                                                                            indicate that this additional post processing stage is effective
                       (a). Flow       Chart of   Fingerprint
                                                                                            in eliminating various types of false minutiae structures.
                                                                                            In combination with the implemented techniques for image
                                                                                            enhancement          and      minutiae         extraction,       preliminary
                                                                                            experiments on the statistics of fingerprints were conducted on
                                                                                            a sample set of fingerprint images. Three types of statistical
                                                                                            data were collected, which include minutiae density, distance
                                                                                            between neighboring minutiae, and ridge wavelength.
                                                                                                  Overall, we have implemented a set of reliable techniques
                                                                                            for    fingerprint    image        enhancement,       minutiae    extraction
         Databue Fingerprint                                                                fingerprint matching and classification. These techniques we
                                                                                            implemented for employer daily attendance system. Through
                                                                                            which employers attend you by fingerprint only enter their
                                                                                            employer ID and put his finger on sensor.

                                                                                                                         V.     CONCLUSION

                                                                                                  Our project     "Biometric       Employer Attendance System
                                                                                            (BEAS)"       is an extensible work for any organization or
                                                                                            company in this fast world. Keeping the view of research still
                                                                        ce)
                       (b). Flow Chart of Fingerprint Matching (Attendan
                                                                                            there is a lot of improvement work and flexibility for the
                Fig.2 ERD Model & Flow Charts of Systems                                    coming technologies in the various demanding directions.
                                                                                            The language which we have is very vast and even the under
D.     Project Evalution
                                                                                            Microsoft products is trying to rule over the Information
     The primary focus of the work in this project is on the
                                                                                            Technology, So we hope that this project will be the point of
enhancement       of      fingerprint     images,      and    the     subsequent
                                                                                            interest for our successors to be enhanced further to market it
extraction of minutiae.
                                                                                            compatible         with     the     demands      of     the    organization
     Firstly, we have implemented a series of techniques for
                                                                                            requirements.
fingerprint image enhancement to facilitate the extraction of
minutiae.      Experiments          were      then     conducted       using    a                                             REFERENCES
combination of both synthetic test images and real fingerprint
                                                                                            [I]  Amengual, J.C, Juan, A, Perz, J.e., Prat, F, SEZ, S., and Villar, J, M,
images in order to provide a well-balanced evaluation on the                                     "Real time minutiae extraction in fingerprint images", proceedings of
performance       of     the implemented algorithm.                 The   use of                 6th International conference on Image Processing and its Applications ,
synthetic images ha provided a more quantative measures of                                       Jully 1997,pp 87 1-875
inspection, but can provide a more realistic evaluation as they                             [2] Dankmeijer, J., Waltman, J.M, and Wild, AG.D "Biological foundations
provide a natural representation of fingerprint imperfections                                   for forensic identification based on fingerprints." Acta Morphological
                                                                                                Neerlando scandivancia 18,1 (1980),67-83
such as noise and corrupted elements. The experimental
                                                                                            [3] Guo, Z, and Hall, R.W "Parallel thinning with two-sub iteration
results have shown that combined with an accurate estimation                                    algorithms", communications of the ACM 32,3 (March 1989),359-373.
of the orientation and ridge frequency; the Gabor filter is able                            [4] Hong, L, Wan, Y and Jain, A.K."Fingerprint image enhancement:
to effectively enhance the clarity of the ridge structures while                                Algorithm and performance evaluation", IEEE transactions on Pattern
                                                                                                Analysis and Machine Intelligence 20, 8(1998),777-789.
reducing noise. In contrast, for low quality images that exhibit
high    intensities      of   noise,    the   filter   is   less    effective   in
enhancing the image due to inaccurate estimation of the




                                                                                     50 1

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:108
posted:7/20/2012
language:
pages:3