Docstoc

Face Recognition Systems (PowerPoint)

Document Sample
Face Recognition Systems (PowerPoint) Powered By Docstoc
					Team-1
Jackie Abbazio
Sasha Perez
Denise Silva
Robert Tesoriero

FACE RECOGNITION SYSTEMS
  Overview: Face Biometrics

 Facial recognition through the use of computer
  analysis of facial structure.
 Software measures a number of points of facial
  characteristics such as eyes, nose, mouth, angles
  of key features, and lengths of various portions.
 Collected data is used to create a template
  through a mathematical algorithm (Neural-
  Networks, Eigenface) and the file is stored within
  the database.
 File is compared to other files within the
  database in search of an identity match.
  Face Biometrics Continued

 Face biometric systems employed the
  capturing of facial pattern characteristics
  through the use of still photography or video
  clips.
 Pattern recognition software relies on:
   Data collection – raw data
   Feature extraction – eyes, nose, mouth, etc
   Classification – class the object is placed into
    (male or female, skin tone, etc)
                FAR & FRR

 FAR ( False Acceptance Rating) – the false
  acceptance rating is the probability that the
  software will incorrectly declare a successful
  match between the input data against the
  database.
 FRR (False Rejection Rating) – the false
  rejection rating is the probability that the
  software will declare a failure to match the
  input data against the database.
                             Our Clients
The clients for team-1 are
Fred Penna, Robert Zack
(DPS ‘10 students), and
Dr. Tappert.
Client/team meetings are
conducted via Skype
conferencing, and are
held on bi-weekly
sessions.
Emails are also
exchanged as per needed
to discuss or schedule
sessions.
Meetings with Dr.
Tappert are conducted
according to progression
issues. Dr Tappert is
contacted at various
times for guidance or
recommendations.
Face Biometric Systems Project

 The face biometric
  systems project will
  incur the research and
  testing of various facial
  biometric software’s. A
  matrix of client
  specifications has been
  given to the team in
  order to find the best
  software to meet the
  client’s needs.
              Matrix of Client
               Specifications
   Software Name
   Website
   Price
   Capture Video
   Extract Features
   Works with these cameras
   Accept these resolutions
   Recognize this Poses
   Works with different lighting
   Source code is available
   Easy to Use Platforms
   Comments
Software Review Matrix
      Luxand’s FaceSDK 1.7

 After extensive testing and researching the
  Face Biometric Systems, we recommend
  purchasing
 Luxand’s FaceSDK 1.7 software. It has the
  following strengths:
   Easy to use
   Ability to enroll all images
   Matches work best at FAR of 50%, but produces
    matches at FAR of 10%.
   Works best for aging
      Luxand FaceSDK Example
                               This 1979 image matched with 2007 image
Enrolled into class database   2008 image with 51% similarity
Matrix 1: Classmates Similarity
             Matrix
Matrix 2: Classmates Facial
Matches
 All face recognition tests were run on
  Luxand’s FaceSDK 1.7, with the FAR set at
  100% and the Minimal Facial Quality at 1.
 By setting FaceSDK 1.7’s options for these
  ratings, we were able produce Similarity
  Ratings and facial matches for all the
  students in the class and provide the data for
  both Matrix 1 & 2 of our presentation.
               Conclusion

 Demonstrations using student photos as the
  database will require FaceSDK’s Similarity
  setting to be at 100% in order for the team to
  build a matrix to involve all students in the
  class.
 By setting the Similarity setting to 100%,
  FaceSDK will give us the ability to compare
  most photos in a numerical rating and a
  visual comparison.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:100
posted:7/22/2011
language:English
pages:13