Access Control Via Face Recognition

brasscoffeeΤεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 3 χρόνια και 10 μήνες)

95 εμφανίσεις

Access Control

Via

Face Recognition

Group Members


Thilanka Priyankara


Vimalaharan Paskarasundaram


Manosha Silva


Dinusha Perera

Introduction


What is Recognition ?



Can you recognize this guy?



Face is belonged to the different guy.



You will feel because your brain is
trained to track such things.


Who is this?

Types of Access Control
Recognition


Finger Print



Shape of Ear



Retina Recognition



Face Recognition

Advantages of Facial Recognition


Give more percentage of guaranty
compared to other techniques.



Intruders should take more effort to
get into the system. (Making a Mask)



With another mechanism, system can
be made to more secure

What we are going to do?


Access Control System



2 way access mechanism

1.
Bar code

2.
Face Recognition



If both mechanisms are satisfied only
user can enter the implemented area







Facial Recognition


We use


1.
Computer Vision

2.
Neural Networks

3.
??????????????




Without Neural Network


If straight forward matching is used


Noise can be there and different lighting


If user comes with a moustache which
he didn’t have before 2 weeks will not be
allowed to get in


Ooh should I have
shave my
moustache to get
in??

Why Neural Network?


Adaptive
-
learning


When users training the system it will keep on learning
about the domain.



Self
-
organization


It gathers the knowledge and organize it by itself



Fault
-
tolerance capabilities


It can tolerate the noise small variations


These features are very much needed to do a system which
can do the human like recognizing

Techniques

Face recognition uses mainly the following
techniques:


Facial geometry:

uses geometrical
characteristics of the face. May use several
cameras to get better accuracy (2D, 3D...)


Skin pattern recognition

(Visual Skin
Print)


Facial thermogram:

uses an infrared
camera to map the face temperatures


Smile:

recognition of the wrinkle changes
when smiling

Facial Geometry


When searching for the facial
features in the face image,



Look for high level features
first


E.g. eyes, nose, mouth


Then relative to high level
features look at the positions of
low level features


E.g. eyebrows

Problems faced, Identifying a Face


find the face in the image


Face finding solves the important task of
making face recognition translation


Scaling of the face image


Distance from the camera


Rotation of the face image if needed


Lighting conditions at the time the
picture is taken

Conclusion


Not easy





Neural network training will eat the
time


Conclusion
cont…


Core technologies are highly researched


Facial scan has unique advantages over
other biometrics


Automated facial detection and facial
recognition algorithm are not yet mature


Facial
-
recognition systems create
opportunities to identify people
unobtrusively