This presentation is based upon the algorithm use in the face and voice detection to help in Biometric process evaluation. We have been researching this topic and could conclude many important aspects which could help readers.
2. Acknowledgement
• I would like to like to thank Mr. David, who gave us this opportunity to
work with Face and Voice biometric cloud detection software. His
ideas are innovative and are kept under Non Disclosure Agreement.
Upon his wish me and my team has made this automated
presentation just to give an overview of what our system is going to
do.
• I would also like to thank my Research and Development team,
without them I could not have successfully completed this
presentation.
3. Our Team of Research and Development
• Team Leader : Bapita Roy
Associate team leader: Soumyajit Maity
R&D engineer: Rudrajit Singh
• Customer Support: Krishna Saha
4. Security using Face and Voice Recognition
There will be two kinds of security system in the project
• Face recognition
• Voice recognition
SSL or https security also be additionally provided for websites
5. Integration of Biometric authentication for applications
The others are Finger print, Hand print, Iris, DNA, Signature etc. These are not part of our
project as of now.
6. Our Face Recognition Algorithm Process
Face detection Face separation
Face recognition
Matches with database
Unlock the System Create Blocks and matches the pattern
7. Algorithm Process
Normal face
2D Alignment
Shape reconstruction Texture Extraction
3D Database
modelsS/W
ExpressionIlluminationPoseRecognition
New face
Classifier
8. The Software will follow the Mathematics
Application reads geometry
of FACE
(as plotted in the grid above)
Points are transferred
to DB as an algorithm
of numbers
This is stored as an
UNIQUE FACE ID by
application
The face ID is mapped with individual details
And user IDENTITY is created.
9. Points of Recognition
Software will assess
the texture of skin
which will detect
moles like
identification
Search for shadows to
identify wrinkles and
age
To identify shape of
lips to identify mood
and gender
Eyebrow shape key to
determine mood of a
person
Any jewelry will help
s/w to identify gender
Shadows cast by hair to
identify age
10. Different features of our face recognition in
our software would be
Permanent protection of your system
Face evolution also it can be detect
Multiple background of images also can be detected
11. Overview to the flow in voice recognition
End point detection Noise Elimination Feature Extraction Voice Model
Voice Matching
Confidence
coefficient
Voice Registration
Result
Voice Identify
Yes
No
No
Voice Verification