Slide 1: Face Recognition Using Neural Networks
Bhavin Pandya EM2007066
Siddhesh Panderkar EM2006044
Gaurav Hansda EM2006022
Hardeepsinh Jadeja EM2006023
Guided By : Prof Hemant Kasturiwale What is Face Recognition? : What is Face Recognition? A face recognition system is a computer application for automatically identifying or verifying a person from a digital image or a video frame from a video source. One of the ways to do this is by comparing selected facial features from the image and a facial database.
Feature to be compared for face recognition:
distance between the lips and the nose
distance between the nose tip and the eyes
distance between the lips and the line joining the two eyes
eccentricity of the face
ratio of the dimensions of the bounding box of the face
width of the lips What are Neural Network? : What are Neural Network? A Neural Network is a system of programs and data structures that approximates the operation of the human brain.
A neural network usually involves a large number of processors operating in parallel, each with its own small sphere of knowledge and access to data in its local memory.
Typically, a neural network is initially "trained" or fed large amounts of data and rules about data relationships (for example, "A grandfather is older than a person's father").
A program can then tell the network how to behave in response to an external stimulus or can initiate activity on its own. MODEL OF NEURON : MODEL OF NEURON Neural Network Architecture : Neural Network Architecture Single layer feed forward network.
Multilayer Feedforward Network
Self Organizing Map(Unsupervised Learning)
Recurrent Network Single layer feedforward network : Single layer feedforward network Multilayer Feedforward Network : Multilayer Feedforward Network Recurrent Networks : Recurrent Networks Learning Algorithms : Learning Algorithms Supervised learning
Reinforcement Learning Approaches to Feature Extraction : Approaches to Feature Extraction Appearance Based
Feature Based (Component Based) Appearance Based Methods : Appearance Based Methods Principle Component Analysis
Linear Discriminant Analysis Block Diagram of Different Training Methods : Block Diagram of Different Training Methods PCA based Face Recognition : PCA based Face Recognition PCA Disadvantages of PCA : 14 Disadvantages of PCA Problems with Eigenfaces (PCA)
Different facial expression Slide 15: Block Diagram of LDA-NN Face Recognition System Steps For Face Recognition Using LDA-NN : Steps For Face Recognition Using LDA-NN Assumptions
Square images with W=H=N
M is the number of images in the database
P is the number of persons in the database Algorithm For LDA-NN Face Recognition. : Algorithm For LDA-NN Face Recognition. The database
We compute the average of all faces
Compute the average face of each person
And subtract them from the training faces Slide 18: We build scatter matrices S1, S2, S3, S4
And the within-class scatter matrix SW
From this scatter matrix we calculate the Fisher face vectors. Fisherfaces, the algorithm : Fisherfaces, the algorithm The database Fisherfaces, the algorithm : Fisherfaces, the algorithm We compute the average of all faces Fisherfaces, the algorithm : Fisherfaces, the algorithm Compute the average face of each person Fisherfaces, the algorithm : Fisherfaces, the algorithm And subtract them from the training faces Fisherfaces, the algorithm : Fisherfaces, the algorithm We build scatter matrices S1, S2, S3, S4
And the within-class scatter matrix SW Slide 24: How is Face Recognition using LDA-NN performed Advantages of LDA-NN : 25 Advantages of LDA-NN Faster than Eigen faces
Has lower error rates
Works well even if different illumination
Works well even if different facial expressions.
Works well with different allignment. Comparison : 26 Comparison FERET database
Best Identification rate: eigenfaces(or PCA) 80.0%, fisherfaces(or LDA) 93.2% Comparison of Different Methods of Face Recognition : Comparison of Different Methods of Face Recognition PROJECT OBJECTIVE : PROJECT OBJECTIVE To implement the concept of Neural Networks for the purpose of Face Recognition.
Further Recognition of unclear images by removing the background noise.
To improve the accuracy of Face recognition by reducing the number of false rejection and false acceptance errors.
To use Face Thermogram that is output of an infrared camera to detect the faces in dark environments.
Recognition of images captured while in motion.
Recognition of faces in videos (motion picture). Advantages : Advantages When an element (Artificial neuron) of the neural network fails, it can continue without any problem by their parallel nature.
A neural network learns and does not need to be reprogrammed.
It can be implemented in any application.
If there is plenty of data and the problem is poorly understood to derive an approximate model, then neural network technology is a good choice. Advantages (contd..) : Advantages (contd..) There is no need to assume an underlying data distribution such as usually is done in statistical modeling.
Neural networks are applicable to multivariate non-linear problems.
The transformations of the variables are automated in the computational process.
A neural network can perform tasks that a linear program can not. Applications of Face Recognition : 31 Applications of Face Recognition Passport control at terminals in airports
Participant identification in meetings
System access control
Scanning for criminal persons Thank you : Thank you