Lattice Computing Extension of the FAM Neural Classifier

gurgleplayΤεχνίτη Νοημοσύνη και Ρομποτική

18 Οκτ 2013 (πριν από 3 χρόνια και 10 μήνες)

73 εμφανίσεις


Lattice Computing Extension of the FAM Neural Classifier
for Human Facial Expression Recognition


Abstract:

A

basic
goal of the human

computer
-
interaction (HCI)

system is to improve the
interactions between users

and computers by making computers more use
r friendly

and receptive to user’s needs. Automatic facial expression

recognition (FER) plays
an important role
. Our
proposes a fundamentally novel

extension
,
flrFAM,

of

the

fuzzy ARTMAP (
FAM
)

classifier for

incremental real
-
time
learning and generalizatio
n based on fuzzy

lattice

reasoning techniques.

FAM

is
enhanced first by a parameter optimization training (sub) phase, and then by a
capacity to process partially ordered (non)numeric data including information
granules.

The

interest here focuses on interv
als' numbers (IN
s)

data, where an IN
represents a
distribution

of

data

samples.

We describe

the

proposed
flrFAM

classifier

as a fuzzy

neural

network that can induce descriptive as well as
flexible (i.e., tunable) decision
-
making knowledge (rules) from

the

data. We
demonstrate

c
apacity

of

the

flrFAM

classifier

for

human

facial

expression

recognition

on benchmark datasets.

The

novel feature extraction as well as
knowledge
-
representation is based on orthogonal moments.

The

reported
experimental results compare

well with

the

results by alternative

classifiers

from

the

literature.

The

far
-
reaching potential

of

fuzzy

lattice

reasoning in

human
-
machine interaction applications is discussed.




Existing System:



Face
Recognition in context (object in a clutter of ot
her objects in the
scene)

and
Object invariance (across different viewpoints, sizes /
distances, & illuminations
.



Face recognition can

recognize faces as a category vs. other objects
.



The homogeneous query addresses the occlusion problem in facial
image re
trieval, while the local feature approach is inadequate.



Using a combination feature values
of

multiple images, we can
compare &
retrieve

co
-
related
faces using only partial information.



Proposed System:




Facial Expression Recognition
familiar faces
virt
ual reality,

computer
games, robotics, machine vision, user profiling for

customer satisfaction
.



T
ensor perceptual color framework

(TPCF) for
Facial Expression
Recognition
(
FER
)

based on information contained in color

facial images,
and investigates perfor
mance
.



The

geometric features present the shape and locations of facial

com
ponents
(including mouth, eyes
, and nose)
.



Optimum features are selected using minimum redundancy maximum
relevance algorithm based on mutual information (IM).



The p
rocessing applic
ations include: perceptual image quality assessment,
face detection, and image segmentation
.





Software Requirement:

Tool : Matlab 2012

Toolbox : Image Processing

Front End: GUI User Interface

Dataset : CAS

Fac
e

Dataset


Hardware R
equirements

SYSTEM


: Pentium IV 2.4 GHz

HARD DISK

: 40 GB

RAM



: 2 GB