SlideShare ist ein Scribd-Unternehmen logo
1 von 18
1

A NOVEL LBP FUZZY FEATURE
EXTRACTION METHOD FOR
FACE RECOGNITION
By
ABDULLAH GUBBI
PA College Of
Engineering
Karnataka

MADASU HANMANDLU
Senior IEEE Member,
EE Dept., IIT Delhi,
New Dehi, India

MOHAMMAD FAZLE AZEEM
Senior IEEE Member,
EE Dept.AMU
UP
Agenda
2

 Face recognition.
 Fuzzy logic.
 Local Binary Pattern.
 Information set.
 K-Nearest Neighbour classifier; Support Vector Machine.
 Results and Conclusion.
3

Face recognition
 Face recognition has received much attention during the past few decades.
 Challenges (I) pose variation (front, non-front), (ii) occlusion, (iii) image
orientation, (iv) illumination condition and (v) facial expression.
 Algorithms (I)Structure-based schemes that make use of shape and other
texture of the face along with 3D depth information.(II) Appearance-based
schemes that make use of the holistic texture features.
 The Eigen faces (PCA) and Fisher faces (LDA) methods are based on the holistic
approach .
4

PCA & LDA
 The Eigen faces (PCA) and Fisher faces (LDA) methods are based on
the holistic approach for face recognition.
 Eigen-faces approach is likely to find the wrong components on
images when there is a large variation in illumination, since the data
points with the maximum variance over all classes are not necessarily
useful for classification.
 The Fisher-faces are computationally expensive.
 Looking at the shortcoming of the above techniques and taking
advantage of Information sets, which have been developed by
Hanmandlu to enlarge the scope of fuzzy sets.
5

Experiment using Haar 10 wavelet de-noising[4]
Recognition
rate

PCA

LDA

ORL database
66.07% 86.07%
(with noise)
Proposed
database
66.43% 89.29%
(without noise)

KPCA

FA

49.29%

85.70%

51.07%

86.07%

Where, PCA- Principle component analysis,
LDA-Linear Discriminant Analysis,
KPCA-Kernel Principle component analysis
and
FA- Fisher Analysis.
6

Local Binary Patterns
 The Local Binary Pattern (LBP) method is widely used in 2D texture
analysis. The LBP operator is a non-parametric 3x3 kernel which
describes the local spatial structure of an image.
 Introduced by Ojala et al.
 LBP is defined as an ordered set of binary comparisons of pixel
intensities between the centre pixel and its eight surrounding pixels.
The decimal values of the resulting 8-bit word (LBP code) leads to
28 possible combinations, which are called Local Binary Patterns
7

Fuzzy Logic
 Fuzzy Logic (FL) theory is the extension of conventional (crisp) set theory.
 It was introduced by Zadeh .
 Deals with the fuzzy sets having imprecise and uncertain data. It handles the
concept of partial truth (truth values between completely true and
completely false.
 Used to model the vagueness and ambiguity in complex systems for which
there is no mathematical model to describe.
 The drawback of fuzzy sets is that they treat the property or attribute values
which we say information source values and their membership function
values separately for all the problems dealing with the fuzzy logic theory.
 consider a set of students graded based on the performance of the class
topper as the benchmark and the student’s individual performance
(information source value) is determined by comparing his performance with
that of the topper (µ).
8

Proposed method
 Instead of considering the whole image and their gray values, divide the
image into sub images of size 3x3 (enclosed in the window)non
overlapping .In case of LBP it is overlapping.
 Extract local information using LBP method

 We compute a membership function for each window based on the
centre pixel of the window using:
ic
window

i( x, y )

i( x, y )
 Where i c represent the centre pixel in the window and
represents the sum of the gray values in the same window.

 The information at the central pixel is given by the product of the
membership value and central pixel value as per the concept of
information value.

Hc

window

ic
9

Proposed method

conti….

 This information is modified by a scale factor = F max with the result the
scaled information is
HS
F max H c
where λ is a scaling parameter which is greater than one (λ>1). and

Fmax

0 . 50

In order to take account of the information from the neighbourhood pixels
(information sources), we compute the LBP value for the window.

Lw

LBP ( x c , y c )

By taking a clue from the communication theory, the information about the
neighbourhood pixels is taken to be the log of the decimal value:

HN

log L w
10

Proposed method

conti….

 Having the information at the central pixel and the neighbourhood
pixels, the total information is taken as the product of these two types
of information, given by

H

HSHN

 The contribution of this method is that it eliminates the shortcoming of
LBP approach that ignores the central pixel value by accounting for
the information of both the central and the neighbourhood pixels
11

Support Vector Machine (SVM)
 A classifier derived from statistical learning theory by Vapnik, et al. in
1992
 SVM became famous when, using images as input, it gave accuracy
comparable to neural-network with hand-designed features in a
handwriting recognition task
 Currently, SVM is widely used in object detection &
recognition, content-based image retrieval, text
recognition, biometrics, speech recognition, etc.
 Also used for regression.

V. Vapnik
12

KNN Classifier
 The k-Nearest Neighbour classifier is amongst the simplest of all the
machine learning methods.
 It is a non-parametric method for classifying objects. Nonparametric in the sense that one need not worry about the
underlying structure.

 Classification is done based on how much close the test feature
vector is to the training feature vectors in the feature space.
 An object is classified based on the majority votes of its neighbours.
If k = 1, then the object is simply assigned to the class of its nearest
neighbour.
13

some of the subjects in databases

 Fig 2. Near-infrared face images of some of the subjects in the CSIST
database.

Fig 3. Gray scale face images of some of the subjects in ORL database

.
Experimental Results
14

The Recognition rates obtained on ORL database with SVM and KNN classifiers

ORL Database

Poly1

Poly 2

KNN

Training images 7
Testing images 3
Training images 3
Testing images 7
Training images 5
Testing images 5
Training images 4
Testing images 6

96.66%

96.66%

92.5%

82.55%

80.35%

82.55%

87.5%

87.5%

88%

88.33%

87.916%

87.08%

Recognition rate for CSIST near infra red database with SVM and KNN classifier.

CSIST Near Infra Red
Database
Training images 1
Testing images 3
Training images 2
Testing images 2

Poly1

Poly 2

KNN

89.33%

87.66%

89.33%

92%

92%

91.55%
15

Flow chart of
implementation

Start
For number of images
Input image train or test
tr Normalize image

Divide into 3x3 windows
For each window
Compute µ, Compute sum
Pick centre pixel and
Compute information set or feature

Store features in database

Stop
16


Conclusion

A novel approach is presented to account for the information from both the central pixel
and the neighbourhood pixels of a face image while matching test sample with the training
samples in the face recognition process .

 The information value is defined as the product of information source value and its
membership function value.


A comparison of performance of the proposed approach is made with that of PCA using
two classifiers KNN and SVM. Better results are reported with SVM.

 The proposed approach is found to be effective on images having variation in
expression, illumination and pose. Further work needs to be done by changing the
membership function values. This is possible one way by employing type-2 membership
functions in which one of the parameters is changed.
17

References

 [1]Laurenz Wiskott, Jean-Marc Fellous, Norbert Kr¨uger, and Christoph von der Malsburg. Face
recognition by elastic bunch graph matching. IEEE Trans. On Pattern Analysis and Machine
Intelligence, 19(7):775–779, 1997.
 [2]M. Turk and A. P. Pentland, “Face Recognition Using Eigenfaces”, IEEE Conference on
Computer Vision and Pattern Recognition, Maui, Hawaii, 1991.

 [3]P. N. Belhumeur, J. P. Hespanha and D. J. Kriegman, “Eigenfaces vs. Fisherfaces: Recognition
Using Class Specific Linear Projection”,IEEE Trans. Pattern Analysis and Machine Intelligence, vol.
19, No. 7, 1997, pp. 711-720.
 [4]Isra’a Abdul-Ameer Wavelet Based Image De-noising to Enhance the Face Recognition Rate,
IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 1, No 3, January 2013.
 [5]Khan, M.M. Face Recognition using Sub-Holistic PCA Information and Communication
Technologies, 2005. ICICT 2005. First International Conference on pages152 – 157( 27-28 Aug. 2005)
 [6]M. Hanmandlu, Information sets and Information Processing, A Research Report, IIT Delhi,
March 2011.
 [7]T. Ojala, M. Pietikäinen, and D. Harwood. A comparative study of texture measures with
classification based on feature distributions. Pattern Recognition, 29, 1996.
 [8]Mamta and M., Hanmandlu, M., Robust Ear Based Authentication Using Local Principal
Independent Components, Expert Systems with Applications, vol. 40, pp. 6478-6490, 2013.
18

THANK YOU

Weitere ähnliche Inhalte

Was ist angesagt?

Facial recognition using modified local binary pattern and random forest
Facial recognition using modified local binary pattern and random forestFacial recognition using modified local binary pattern and random forest
Facial recognition using modified local binary pattern and random forestijaia
 
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...IJCSEIT Journal
 
Implementation of High Dimension Colour Transform in Domain of Image Processing
Implementation of High Dimension Colour Transform in Domain of Image ProcessingImplementation of High Dimension Colour Transform in Domain of Image Processing
Implementation of High Dimension Colour Transform in Domain of Image ProcessingIRJET Journal
 
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...CSCJournals
 
A Novel Feature Extraction Scheme for Medical X-Ray Images
A Novel Feature Extraction Scheme for Medical X-Ray ImagesA Novel Feature Extraction Scheme for Medical X-Ray Images
A Novel Feature Extraction Scheme for Medical X-Ray ImagesIJERA Editor
 
Face Image Restoration based on sample patterns using MLP Neural Network
Face Image Restoration based on sample patterns using MLP Neural NetworkFace Image Restoration based on sample patterns using MLP Neural Network
Face Image Restoration based on sample patterns using MLP Neural NetworkIOSR Journals
 
Performance Evaluation of Illumination Invariant Face Recognition Algorthims
Performance Evaluation of Illumination Invariant Face Recognition AlgorthimsPerformance Evaluation of Illumination Invariant Face Recognition Algorthims
Performance Evaluation of Illumination Invariant Face Recognition AlgorthimsIRJET Journal
 
Face skin color based recognition using local spectral and gray scale features
Face skin color based recognition using local spectral and gray scale featuresFace skin color based recognition using local spectral and gray scale features
Face skin color based recognition using local spectral and gray scale featureseSAT Journals
 
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLING
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLINGAPPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLING
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLINGsipij
 
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERN
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERNOFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERN
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERNijaia
 
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual Learning
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual LearningSequential Reptile_Inter-Task Gradient Alignment for Multilingual Learning
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual LearningMLAI2
 
Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation
Fuzzy Region Merging Using Fuzzy Similarity Measurement  on Image Segmentation  Fuzzy Region Merging Using Fuzzy Similarity Measurement  on Image Segmentation
Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation IJECEIAES
 
Face Detection for identification of people in Images of Internet
Face Detection for identification of people in Images of InternetFace Detection for identification of people in Images of Internet
Face Detection for identification of people in Images of Internetijceronline
 
Introduction to image processing and pattern recognition
Introduction to image processing and pattern recognitionIntroduction to image processing and pattern recognition
Introduction to image processing and pattern recognitionSaibee Alam
 
Improvement of Objective Image Quality Evaluation Applying Colour Differences...
Improvement of Objective Image Quality Evaluation Applying Colour Differences...Improvement of Objective Image Quality Evaluation Applying Colour Differences...
Improvement of Objective Image Quality Evaluation Applying Colour Differences...CSCJournals
 
Ajay ppt region segmentation new copy
Ajay ppt region segmentation new   copyAjay ppt region segmentation new   copy
Ajay ppt region segmentation new copyAjay Kumar Singh
 

Was ist angesagt? (19)

Facial recognition using modified local binary pattern and random forest
Facial recognition using modified local binary pattern and random forestFacial recognition using modified local binary pattern and random forest
Facial recognition using modified local binary pattern and random forest
 
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...
FACE RECOGNITION USING DIFFERENT LOCAL FEATURES WITH DIFFERENT DISTANCE TECHN...
 
Implementation of High Dimension Colour Transform in Domain of Image Processing
Implementation of High Dimension Colour Transform in Domain of Image ProcessingImplementation of High Dimension Colour Transform in Domain of Image Processing
Implementation of High Dimension Colour Transform in Domain of Image Processing
 
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...
Shallow vs. Deep Image Representations: A Comparative Study with Enhancements...
 
A Novel Feature Extraction Scheme for Medical X-Ray Images
A Novel Feature Extraction Scheme for Medical X-Ray ImagesA Novel Feature Extraction Scheme for Medical X-Ray Images
A Novel Feature Extraction Scheme for Medical X-Ray Images
 
H017416670
H017416670H017416670
H017416670
 
Face Image Restoration based on sample patterns using MLP Neural Network
Face Image Restoration based on sample patterns using MLP Neural NetworkFace Image Restoration based on sample patterns using MLP Neural Network
Face Image Restoration based on sample patterns using MLP Neural Network
 
Performance Evaluation of Illumination Invariant Face Recognition Algorthims
Performance Evaluation of Illumination Invariant Face Recognition AlgorthimsPerformance Evaluation of Illumination Invariant Face Recognition Algorthims
Performance Evaluation of Illumination Invariant Face Recognition Algorthims
 
Face skin color based recognition using local spectral and gray scale features
Face skin color based recognition using local spectral and gray scale featuresFace skin color based recognition using local spectral and gray scale features
Face skin color based recognition using local spectral and gray scale features
 
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLING
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLINGAPPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLING
APPLYING R-SPATIOGRAM IN OBJECT TRACKING FOR OCCLUSION HANDLING
 
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERN
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERNOFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERN
OFFLINE SIGNATURE VERIFICATION USING SUPPORT LOCAL BINARY PATTERN
 
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual Learning
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual LearningSequential Reptile_Inter-Task Gradient Alignment for Multilingual Learning
Sequential Reptile_Inter-Task Gradient Alignment for Multilingual Learning
 
Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation
Fuzzy Region Merging Using Fuzzy Similarity Measurement  on Image Segmentation  Fuzzy Region Merging Using Fuzzy Similarity Measurement  on Image Segmentation
Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation
 
Face Detection for identification of people in Images of Internet
Face Detection for identification of people in Images of InternetFace Detection for identification of people in Images of Internet
Face Detection for identification of people in Images of Internet
 
Introduction to image processing and pattern recognition
Introduction to image processing and pattern recognitionIntroduction to image processing and pattern recognition
Introduction to image processing and pattern recognition
 
C1104011322
C1104011322C1104011322
C1104011322
 
Improvement of Objective Image Quality Evaluation Applying Colour Differences...
Improvement of Objective Image Quality Evaluation Applying Colour Differences...Improvement of Objective Image Quality Evaluation Applying Colour Differences...
Improvement of Objective Image Quality Evaluation Applying Colour Differences...
 
B018110915
B018110915B018110915
B018110915
 
Ajay ppt region segmentation new copy
Ajay ppt region segmentation new   copyAjay ppt region segmentation new   copy
Ajay ppt region segmentation new copy
 

Andere mochten auch

Facial expression recognition based on local binary patterns final
Facial expression recognition based on local binary patterns finalFacial expression recognition based on local binary patterns final
Facial expression recognition based on local binary patterns finalahmad abdelhafeez
 
FACE RECOGNITION TECHNOLOGY
FACE RECOGNITION TECHNOLOGYFACE RECOGNITION TECHNOLOGY
FACE RECOGNITION TECHNOLOGYJASHU JASWANTH
 
Face recognition technology - BEST PPT
Face recognition technology - BEST PPTFace recognition technology - BEST PPT
Face recognition technology - BEST PPTSiddharth Modi
 
Face recognition ppt
Face recognition pptFace recognition ppt
Face recognition pptSantosh Kumar
 
讀書會 Fuzzy在人臉辨識上的應用
讀書會 Fuzzy在人臉辨識上的應用讀書會 Fuzzy在人臉辨識上的應用
讀書會 Fuzzy在人臉辨識上的應用Makris Huang
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technologyPushkar Dutt
 
December 2007 Document Recognition Technology Overview Presentation
December 2007 Document Recognition Technology Overview PresentationDecember 2007 Document Recognition Technology Overview Presentation
December 2007 Document Recognition Technology Overview PresentationJohn Wang
 
4 g technology
4 g  technology4 g  technology
4 g technologyRitu Bafna
 
Single Sample Soft Shadows Using Depth Maps
Single Sample Soft Shadows Using Depth MapsSingle Sample Soft Shadows Using Depth Maps
Single Sample Soft Shadows Using Depth Mapsstefan_b
 
Robust Human Emotion Analysis Using LBP, GLCM and PNN Classifier
Robust Human Emotion Analysis Using LBP, GLCM and PNN ClassifierRobust Human Emotion Analysis Using LBP, GLCM and PNN Classifier
Robust Human Emotion Analysis Using LBP, GLCM and PNN ClassifierIJTET Journal
 
Expression invariant face recognition
Expression invariant face recognitionExpression invariant face recognition
Expression invariant face recognitionSumit Agarwal
 
Advanced Support Vector Machine for classification in Neural Network
Advanced Support Vector Machine for classification  in Neural NetworkAdvanced Support Vector Machine for classification  in Neural Network
Advanced Support Vector Machine for classification in Neural NetworkAshwani Jha
 
Multiple Kernel Learning based Approach to Representation and Feature Selecti...
Multiple Kernel Learning based Approach to Representation and Feature Selecti...Multiple Kernel Learning based Approach to Representation and Feature Selecti...
Multiple Kernel Learning based Approach to Representation and Feature Selecti...ICAC09
 
Document Recognition Technologies
Document Recognition TechnologiesDocument Recognition Technologies
Document Recognition TechnologiesChris Riley ☁
 
PERFORMANCE EVALUATION AND IMPLEMENTATION OF FACIAL EXPRESSION AND EMOTION R...
PERFORMANCE EVALUATION AND IMPLEMENTATION  OF FACIAL EXPRESSION AND EMOTION R...PERFORMANCE EVALUATION AND IMPLEMENTATION  OF FACIAL EXPRESSION AND EMOTION R...
PERFORMANCE EVALUATION AND IMPLEMENTATION OF FACIAL EXPRESSION AND EMOTION R...Akhil Upadhyay
 
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...vnktrjr
 
2.6 support vector machines and associative classifiers revised
2.6 support vector machines and associative classifiers revised2.6 support vector machines and associative classifiers revised
2.6 support vector machines and associative classifiers revisedKrish_ver2
 

Andere mochten auch (20)

Facial expression recognition based on local binary patterns final
Facial expression recognition based on local binary patterns finalFacial expression recognition based on local binary patterns final
Facial expression recognition based on local binary patterns final
 
FACE RECOGNITION TECHNOLOGY
FACE RECOGNITION TECHNOLOGYFACE RECOGNITION TECHNOLOGY
FACE RECOGNITION TECHNOLOGY
 
Face recognition technology - BEST PPT
Face recognition technology - BEST PPTFace recognition technology - BEST PPT
Face recognition technology - BEST PPT
 
Face recognition ppt
Face recognition pptFace recognition ppt
Face recognition ppt
 
讀書會 Fuzzy在人臉辨識上的應用
讀書會 Fuzzy在人臉辨識上的應用讀書會 Fuzzy在人臉辨識上的應用
讀書會 Fuzzy在人臉辨識上的應用
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technology
 
December 2007 Document Recognition Technology Overview Presentation
December 2007 Document Recognition Technology Overview PresentationDecember 2007 Document Recognition Technology Overview Presentation
December 2007 Document Recognition Technology Overview Presentation
 
4 g technology
4 g  technology4 g  technology
4 g technology
 
Single Sample Soft Shadows Using Depth Maps
Single Sample Soft Shadows Using Depth MapsSingle Sample Soft Shadows Using Depth Maps
Single Sample Soft Shadows Using Depth Maps
 
Robust Human Emotion Analysis Using LBP, GLCM and PNN Classifier
Robust Human Emotion Analysis Using LBP, GLCM and PNN ClassifierRobust Human Emotion Analysis Using LBP, GLCM and PNN Classifier
Robust Human Emotion Analysis Using LBP, GLCM and PNN Classifier
 
Expression invariant face recognition
Expression invariant face recognitionExpression invariant face recognition
Expression invariant face recognition
 
Advanced Support Vector Machine for classification in Neural Network
Advanced Support Vector Machine for classification  in Neural NetworkAdvanced Support Vector Machine for classification  in Neural Network
Advanced Support Vector Machine for classification in Neural Network
 
Multiple Kernel Learning based Approach to Representation and Feature Selecti...
Multiple Kernel Learning based Approach to Representation and Feature Selecti...Multiple Kernel Learning based Approach to Representation and Feature Selecti...
Multiple Kernel Learning based Approach to Representation and Feature Selecti...
 
C3.4.2
C3.4.2C3.4.2
C3.4.2
 
Document Recognition Technologies
Document Recognition TechnologiesDocument Recognition Technologies
Document Recognition Technologies
 
PERFORMANCE EVALUATION AND IMPLEMENTATION OF FACIAL EXPRESSION AND EMOTION R...
PERFORMANCE EVALUATION AND IMPLEMENTATION  OF FACIAL EXPRESSION AND EMOTION R...PERFORMANCE EVALUATION AND IMPLEMENTATION  OF FACIAL EXPRESSION AND EMOTION R...
PERFORMANCE EVALUATION AND IMPLEMENTATION OF FACIAL EXPRESSION AND EMOTION R...
 
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...
STUDY ON IMPROVED RADIATION PERFORMANCE CHARACTERISTICS OF FRACTAL ANTENNA FO...
 
BC Workflow
BC WorkflowBC Workflow
BC Workflow
 
2.6 support vector machines and associative classifiers revised
2.6 support vector machines and associative classifiers revised2.6 support vector machines and associative classifiers revised
2.6 support vector machines and associative classifiers revised
 
Ijariie1177
Ijariie1177Ijariie1177
Ijariie1177
 

Ähnlich wie fuzzy LBP for face recognition ppt

Selective local binary pattern with convolutional neural network for facial ...
Selective local binary pattern with convolutional neural  network for facial ...Selective local binary pattern with convolutional neural  network for facial ...
Selective local binary pattern with convolutional neural network for facial ...IJECEIAES
 
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVM
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVMCOMPRESSION BASED FACE RECOGNITION USING DWT AND SVM
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVMsipij
 
最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - 最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - Hiroshi Fukui
 
An ensemble classification algorithm for hyperspectral images
An ensemble classification algorithm for hyperspectral imagesAn ensemble classification algorithm for hyperspectral images
An ensemble classification algorithm for hyperspectral imagessipij
 
Face Recognition Using Neural Networks
Face Recognition Using Neural NetworksFace Recognition Using Neural Networks
Face Recognition Using Neural NetworksCSCJournals
 
Human Re-identification with Global and Local Siamese Convolution Neural Network
Human Re-identification with Global and Local Siamese Convolution Neural NetworkHuman Re-identification with Global and Local Siamese Convolution Neural Network
Human Re-identification with Global and Local Siamese Convolution Neural NetworkTELKOMNIKA JOURNAL
 
Rigorous Pack Edge Detection Fuzzy System
Rigorous Pack Edge Detection Fuzzy SystemRigorous Pack Edge Detection Fuzzy System
Rigorous Pack Edge Detection Fuzzy Systeminventy
 
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...inventionjournals
 
Face Recognition Using Gabor features And PCA
Face Recognition Using Gabor features And PCAFace Recognition Using Gabor features And PCA
Face Recognition Using Gabor features And PCAIOSR Journals
 
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...TELKOMNIKA JOURNAL
 
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDY
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDYSINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDY
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDYcsandit
 
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...ijma
 
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural Network
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural NetworkTargeted Visual Content Recognition Using Multi-Layer Perceptron Neural Network
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural Networkijceronline
 
Deep learning for pose-invariant face detection in unconstrained environment
Deep learning for pose-invariant face detection in unconstrained environmentDeep learning for pose-invariant face detection in unconstrained environment
Deep learning for pose-invariant face detection in unconstrained environmentIJECEIAES
 
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION sipij
 
Face Identification Project Abstract 2017
Face Identification Project Abstract 2017Face Identification Project Abstract 2017
Face Identification Project Abstract 2017ioshean
 

Ähnlich wie fuzzy LBP for face recognition ppt (20)

Selective local binary pattern with convolutional neural network for facial ...
Selective local binary pattern with convolutional neural  network for facial ...Selective local binary pattern with convolutional neural  network for facial ...
Selective local binary pattern with convolutional neural network for facial ...
 
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVM
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVMCOMPRESSION BASED FACE RECOGNITION USING DWT AND SVM
COMPRESSION BASED FACE RECOGNITION USING DWT AND SVM
 
最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に - 最近の研究情勢についていくために - Deep Learningを中心に -
最近の研究情勢についていくために - Deep Learningを中心に -
 
An ensemble classification algorithm for hyperspectral images
An ensemble classification algorithm for hyperspectral imagesAn ensemble classification algorithm for hyperspectral images
An ensemble classification algorithm for hyperspectral images
 
Face Recognition Using Neural Networks
Face Recognition Using Neural NetworksFace Recognition Using Neural Networks
Face Recognition Using Neural Networks
 
Human Re-identification with Global and Local Siamese Convolution Neural Network
Human Re-identification with Global and Local Siamese Convolution Neural NetworkHuman Re-identification with Global and Local Siamese Convolution Neural Network
Human Re-identification with Global and Local Siamese Convolution Neural Network
 
Rigorous Pack Edge Detection Fuzzy System
Rigorous Pack Edge Detection Fuzzy SystemRigorous Pack Edge Detection Fuzzy System
Rigorous Pack Edge Detection Fuzzy System
 
Dx25751756
Dx25751756Dx25751756
Dx25751756
 
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...
Face Recognition System Using Local Ternary Pattern and Signed Number Multipl...
 
H0334749
H0334749H0334749
H0334749
 
Face Recognition Using Gabor features And PCA
Face Recognition Using Gabor features And PCAFace Recognition Using Gabor features And PCA
Face Recognition Using Gabor features And PCA
 
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...
A Crop Pests Image Classification Algorithm Based on Deep Convolutional Neura...
 
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDY
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDYSINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDY
SINGLE IMAGE SUPER RESOLUTION: A COMPARATIVE STUDY
 
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...
Multilinear Kernel Mapping for Feature Dimension Reduction in Content Based M...
 
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural Network
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural NetworkTargeted Visual Content Recognition Using Multi-Layer Perceptron Neural Network
Targeted Visual Content Recognition Using Multi-Layer Perceptron Neural Network
 
Deep learning for pose-invariant face detection in unconstrained environment
Deep learning for pose-invariant face detection in unconstrained environmentDeep learning for pose-invariant face detection in unconstrained environment
Deep learning for pose-invariant face detection in unconstrained environment
 
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION
HVDLP : HORIZONTAL VERTICAL DIAGONAL LOCAL PATTERN BASED FACE RECOGNITION
 
F017533540
F017533540F017533540
F017533540
 
Face Identification Project Abstract 2017
Face Identification Project Abstract 2017Face Identification Project Abstract 2017
Face Identification Project Abstract 2017
 
S0450598102
S0450598102S0450598102
S0450598102
 

Kürzlich hochgeladen

Web & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfWeb & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfJayanti Pande
 
Grant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy ConsultingGrant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy ConsultingTechSoup
 
Measures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDMeasures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDThiyagu K
 
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)eniolaolutunde
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxheathfieldcps1
 
9548086042 for call girls in Indira Nagar with room service
9548086042  for call girls in Indira Nagar  with room service9548086042  for call girls in Indira Nagar  with room service
9548086042 for call girls in Indira Nagar with room servicediscovermytutordmt
 
Introduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsIntroduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsTechSoup
 
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...fonyou31
 
Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Disha Kariya
 
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdfBASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdfSoniaTolstoy
 
The Most Excellent Way | 1 Corinthians 13
The Most Excellent Way | 1 Corinthians 13The Most Excellent Way | 1 Corinthians 13
The Most Excellent Way | 1 Corinthians 13Steve Thomason
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfciinovamais
 
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...PsychoTech Services
 
Q4-W6-Restating Informational Text Grade 3
Q4-W6-Restating Informational Text Grade 3Q4-W6-Restating Informational Text Grade 3
Q4-W6-Restating Informational Text Grade 3JemimahLaneBuaron
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityGeoBlogs
 
Student login on Anyboli platform.helpin
Student login on Anyboli platform.helpinStudent login on Anyboli platform.helpin
Student login on Anyboli platform.helpinRaunakKeshri1
 
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...Krashi Coaching
 
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...Sapna Thakur
 

Kürzlich hochgeladen (20)

Web & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdfWeb & Social Media Analytics Previous Year Question Paper.pdf
Web & Social Media Analytics Previous Year Question Paper.pdf
 
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
Mattingly "AI & Prompt Design: Structured Data, Assistants, & RAG"
 
Grant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy ConsultingGrant Readiness 101 TechSoup and Remy Consulting
Grant Readiness 101 TechSoup and Remy Consulting
 
Measures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SDMeasures of Dispersion and Variability: Range, QD, AD and SD
Measures of Dispersion and Variability: Range, QD, AD and SD
 
Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptx
 
9548086042 for call girls in Indira Nagar with room service
9548086042  for call girls in Indira Nagar  with room service9548086042  for call girls in Indira Nagar  with room service
9548086042 for call girls in Indira Nagar with room service
 
Introduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsIntroduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The Basics
 
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...
Ecosystem Interactions Class Discussion Presentation in Blue Green Lined Styl...
 
Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..Sports & Fitness Value Added Course FY..
Sports & Fitness Value Added Course FY..
 
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdfBASLIQ CURRENT LOOKBOOK  LOOKBOOK(1) (1).pdf
BASLIQ CURRENT LOOKBOOK LOOKBOOK(1) (1).pdf
 
The Most Excellent Way | 1 Corinthians 13
The Most Excellent Way | 1 Corinthians 13The Most Excellent Way | 1 Corinthians 13
The Most Excellent Way | 1 Corinthians 13
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...
IGNOU MSCCFT and PGDCFT Exam Question Pattern: MCFT003 Counselling and Family...
 
Mattingly "AI & Prompt Design: The Basics of Prompt Design"
Mattingly "AI & Prompt Design: The Basics of Prompt Design"Mattingly "AI & Prompt Design: The Basics of Prompt Design"
Mattingly "AI & Prompt Design: The Basics of Prompt Design"
 
Q4-W6-Restating Informational Text Grade 3
Q4-W6-Restating Informational Text Grade 3Q4-W6-Restating Informational Text Grade 3
Q4-W6-Restating Informational Text Grade 3
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activity
 
Student login on Anyboli platform.helpin
Student login on Anyboli platform.helpinStudent login on Anyboli platform.helpin
Student login on Anyboli platform.helpin
 
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
Kisan Call Centre - To harness potential of ICT in Agriculture by answer farm...
 
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...
BAG TECHNIQUE Bag technique-a tool making use of public health bag through wh...
 

fuzzy LBP for face recognition ppt

  • 1. 1 A NOVEL LBP FUZZY FEATURE EXTRACTION METHOD FOR FACE RECOGNITION By ABDULLAH GUBBI PA College Of Engineering Karnataka MADASU HANMANDLU Senior IEEE Member, EE Dept., IIT Delhi, New Dehi, India MOHAMMAD FAZLE AZEEM Senior IEEE Member, EE Dept.AMU UP
  • 2. Agenda 2  Face recognition.  Fuzzy logic.  Local Binary Pattern.  Information set.  K-Nearest Neighbour classifier; Support Vector Machine.  Results and Conclusion.
  • 3. 3 Face recognition  Face recognition has received much attention during the past few decades.  Challenges (I) pose variation (front, non-front), (ii) occlusion, (iii) image orientation, (iv) illumination condition and (v) facial expression.  Algorithms (I)Structure-based schemes that make use of shape and other texture of the face along with 3D depth information.(II) Appearance-based schemes that make use of the holistic texture features.  The Eigen faces (PCA) and Fisher faces (LDA) methods are based on the holistic approach .
  • 4. 4 PCA & LDA  The Eigen faces (PCA) and Fisher faces (LDA) methods are based on the holistic approach for face recognition.  Eigen-faces approach is likely to find the wrong components on images when there is a large variation in illumination, since the data points with the maximum variance over all classes are not necessarily useful for classification.  The Fisher-faces are computationally expensive.  Looking at the shortcoming of the above techniques and taking advantage of Information sets, which have been developed by Hanmandlu to enlarge the scope of fuzzy sets.
  • 5. 5 Experiment using Haar 10 wavelet de-noising[4] Recognition rate PCA LDA ORL database 66.07% 86.07% (with noise) Proposed database 66.43% 89.29% (without noise) KPCA FA 49.29% 85.70% 51.07% 86.07% Where, PCA- Principle component analysis, LDA-Linear Discriminant Analysis, KPCA-Kernel Principle component analysis and FA- Fisher Analysis.
  • 6. 6 Local Binary Patterns  The Local Binary Pattern (LBP) method is widely used in 2D texture analysis. The LBP operator is a non-parametric 3x3 kernel which describes the local spatial structure of an image.  Introduced by Ojala et al.  LBP is defined as an ordered set of binary comparisons of pixel intensities between the centre pixel and its eight surrounding pixels. The decimal values of the resulting 8-bit word (LBP code) leads to 28 possible combinations, which are called Local Binary Patterns
  • 7. 7 Fuzzy Logic  Fuzzy Logic (FL) theory is the extension of conventional (crisp) set theory.  It was introduced by Zadeh .  Deals with the fuzzy sets having imprecise and uncertain data. It handles the concept of partial truth (truth values between completely true and completely false.  Used to model the vagueness and ambiguity in complex systems for which there is no mathematical model to describe.  The drawback of fuzzy sets is that they treat the property or attribute values which we say information source values and their membership function values separately for all the problems dealing with the fuzzy logic theory.  consider a set of students graded based on the performance of the class topper as the benchmark and the student’s individual performance (information source value) is determined by comparing his performance with that of the topper (µ).
  • 8. 8 Proposed method  Instead of considering the whole image and their gray values, divide the image into sub images of size 3x3 (enclosed in the window)non overlapping .In case of LBP it is overlapping.  Extract local information using LBP method  We compute a membership function for each window based on the centre pixel of the window using: ic window i( x, y ) i( x, y )  Where i c represent the centre pixel in the window and represents the sum of the gray values in the same window.  The information at the central pixel is given by the product of the membership value and central pixel value as per the concept of information value. Hc window ic
  • 9. 9 Proposed method conti….  This information is modified by a scale factor = F max with the result the scaled information is HS F max H c where λ is a scaling parameter which is greater than one (λ>1). and Fmax 0 . 50 In order to take account of the information from the neighbourhood pixels (information sources), we compute the LBP value for the window. Lw LBP ( x c , y c ) By taking a clue from the communication theory, the information about the neighbourhood pixels is taken to be the log of the decimal value: HN log L w
  • 10. 10 Proposed method conti….  Having the information at the central pixel and the neighbourhood pixels, the total information is taken as the product of these two types of information, given by H HSHN  The contribution of this method is that it eliminates the shortcoming of LBP approach that ignores the central pixel value by accounting for the information of both the central and the neighbourhood pixels
  • 11. 11 Support Vector Machine (SVM)  A classifier derived from statistical learning theory by Vapnik, et al. in 1992  SVM became famous when, using images as input, it gave accuracy comparable to neural-network with hand-designed features in a handwriting recognition task  Currently, SVM is widely used in object detection & recognition, content-based image retrieval, text recognition, biometrics, speech recognition, etc.  Also used for regression. V. Vapnik
  • 12. 12 KNN Classifier  The k-Nearest Neighbour classifier is amongst the simplest of all the machine learning methods.  It is a non-parametric method for classifying objects. Nonparametric in the sense that one need not worry about the underlying structure.  Classification is done based on how much close the test feature vector is to the training feature vectors in the feature space.  An object is classified based on the majority votes of its neighbours. If k = 1, then the object is simply assigned to the class of its nearest neighbour.
  • 13. 13 some of the subjects in databases  Fig 2. Near-infrared face images of some of the subjects in the CSIST database. Fig 3. Gray scale face images of some of the subjects in ORL database .
  • 14. Experimental Results 14 The Recognition rates obtained on ORL database with SVM and KNN classifiers ORL Database Poly1 Poly 2 KNN Training images 7 Testing images 3 Training images 3 Testing images 7 Training images 5 Testing images 5 Training images 4 Testing images 6 96.66% 96.66% 92.5% 82.55% 80.35% 82.55% 87.5% 87.5% 88% 88.33% 87.916% 87.08% Recognition rate for CSIST near infra red database with SVM and KNN classifier. CSIST Near Infra Red Database Training images 1 Testing images 3 Training images 2 Testing images 2 Poly1 Poly 2 KNN 89.33% 87.66% 89.33% 92% 92% 91.55%
  • 15. 15 Flow chart of implementation Start For number of images Input image train or test tr Normalize image Divide into 3x3 windows For each window Compute µ, Compute sum Pick centre pixel and Compute information set or feature Store features in database Stop
  • 16. 16  Conclusion A novel approach is presented to account for the information from both the central pixel and the neighbourhood pixels of a face image while matching test sample with the training samples in the face recognition process .  The information value is defined as the product of information source value and its membership function value.  A comparison of performance of the proposed approach is made with that of PCA using two classifiers KNN and SVM. Better results are reported with SVM.  The proposed approach is found to be effective on images having variation in expression, illumination and pose. Further work needs to be done by changing the membership function values. This is possible one way by employing type-2 membership functions in which one of the parameters is changed.
  • 17. 17 References  [1]Laurenz Wiskott, Jean-Marc Fellous, Norbert Kr¨uger, and Christoph von der Malsburg. Face recognition by elastic bunch graph matching. IEEE Trans. On Pattern Analysis and Machine Intelligence, 19(7):775–779, 1997.  [2]M. Turk and A. P. Pentland, “Face Recognition Using Eigenfaces”, IEEE Conference on Computer Vision and Pattern Recognition, Maui, Hawaii, 1991.  [3]P. N. Belhumeur, J. P. Hespanha and D. J. Kriegman, “Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection”,IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, No. 7, 1997, pp. 711-720.  [4]Isra’a Abdul-Ameer Wavelet Based Image De-noising to Enhance the Face Recognition Rate, IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 1, No 3, January 2013.  [5]Khan, M.M. Face Recognition using Sub-Holistic PCA Information and Communication Technologies, 2005. ICICT 2005. First International Conference on pages152 – 157( 27-28 Aug. 2005)  [6]M. Hanmandlu, Information sets and Information Processing, A Research Report, IIT Delhi, March 2011.  [7]T. Ojala, M. Pietikäinen, and D. Harwood. A comparative study of texture measures with classification based on feature distributions. Pattern Recognition, 29, 1996.  [8]Mamta and M., Hanmandlu, M., Robust Ear Based Authentication Using Local Principal Independent Components, Expert Systems with Applications, vol. 40, pp. 6478-6490, 2013.