Face Recognition

crumcasteΤεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

72 εμφανίσεις

Face Recognition

eScience Project of MIT

Comp 6702 (18 units)

Introduction


Client:

Professor Tom Gedeon


Project Purpose:


1
)

Propose

some

new

idea

or

do

some

enforcement

on

current

Face

Recognition

Technology

(FRT)
.



2
)

Do

comparisons

among

our

ideas

and

other

existing

method

and

analysis

the

comparison

result




Introduction of FRT

What is it about?


The idea of FRT is to
use technology to
achieve the goal of
the natural
mechanisms of the
human visual system.

Why do we need it?


Non
-
intrusive nature



Extremely useful for
the situation such as
security, tracing the
criminals and so on

Problems


There are still a lot of restrictions on
how the facial images are obtained.




The result may be affected greatly from
the variances in angles, expressions,
illuminations and so on
.

Our ideas

Trigram in FRT


Find the patterns from
the face image.


Use small parts of the
images as if they were
words in a text file


The process of
identify a face is to
find the images with
similar pattern


PCA Channel


PCA method


Limitation of PCA
methods


PCA Channels


General Plan


Implement various FRTs, such as PCA,
LDA and including the two we proposed.


Implement Neuro Network


Do comparisons among the Algorithms


Analysis the results and deliver the
analysis report or paper


Finalize the software left from last
semester


Integrate all the FRTs into one system


Timetable