Face Recognition

15
Face Recognition using Direct Fractional Step Linear Discriminant Algorithm (DF-LDA) Mohammad Anwar Sumit Sampat Mohammad Murtuza

Transcript of Face Recognition

Page 1: Face Recognition

Face Recognition

using Direct Fractional Step Linear Discriminant Algorithm (DF-LDA)

Mohammad AnwarSumit Sampat

Mohammad Murtuza

Page 2: Face Recognition

Progress

Understanding of related material is completed.

Beginning of implementation has begun.

Page 3: Face Recognition

DF-LDA Algorithm

The output are the Optimal Discriminant Features (ODF) which willbe projected onto the DF-LDA based subspace.

(size of the image)

Page 4: Face Recognition

DF-LDA Algorithm – contd.

= between class scatter matrix and iswhere

• choice of weighting function - Euclidean

Page 5: Face Recognition

DF-LDA Algorithm – contd.

diagonal matrix.

> 0

Page 6: Face Recognition

DF-LDA Algorithm – contd.

Since we need to maximize the ratio

Page 7: Face Recognition

DF-LDA Algorithm – contd.

The basic idea of complex mathematical computations in orderTo reduce the dimensions is to obtain better classification. TheF-LDA step is incorporated in the case of closed classes to obtain the required ODF’s.

Page 8: Face Recognition

Feret Dataset

Feret Dataset consists of facial images at different angles which are not normalized. Also includes data files which have feature data points (ground truths). Feature data points – includes x, y-axis locations

of facial features such as left and right eyes, nose, and center of mouth.

Page 9: Face Recognition

Feret Image Example

00001fa010_930831.tif

history= FERET;../00001fa010.930831date_taken= 31 Aug 1993eye_glasses= yespose= frontallighting= inside_feretmedia= film_still ASA_200expression= FAid= 00001image_processing= histogram_adjustedimage_resize= 0%image_bright_reduction= 0%feret_pose_flag= faleft_eye_coords= 163 166right_eye_coords= 101 167nose_tip_coords= 134 202mouth_center_coords= 133 234

00001fa010_930831.gnd

Page 10: Face Recognition

Implementation preprocessing image files

Each image in Feret dataset are bzipped and and in tiff image format.

Using a simple batch script to unzip the files and convert tiff image files to PGM format using convert function from ImageMagick 5.5.7.

URL for ImageMagick www.imagemagick.org

Page 11: Face Recognition

Implementation creating eye-coordinates file

Each image in Feret has an associated .gnd file which lists feature data points.

Perl script is used to extract left and right eye coordinates from each .gnd file and placed into an one eye coordinate file which lists image name and eye-coordinates.

00001fa010_930831.pgm 163 166 101 16700002fa010_930831.pgm 135 136 100 18900003fa010_930831.pgm 168 169 112 170

Eye coordinates file

Page 12: Face Recognition

Implementation normalization

Normalization is done by reading each line in the eye coordinates file. These are the steps for normalization [4]:

The image is scaled so as to make the distance between the eye's constant. In this step, the image is also cropped to a smaller size that will include essentially just the face. The standard FERET normalization crops the image to 150x130 pixels with 70 pixels between the centers of the eyes.

A mask is applied that zeroes out pixels not in an oval that contains the typical face. Thus, hair, shirt collars, etc. are typically removed. The mask is generated analytically by specifying the dimensions of the masking oval.

Histogram equalization is used to smooth the distribution of grey values for the non-masked pixels.

The image is normalized so the non-masked pixels have mean zero and standard deviation one.

00001fa010_930831.pgm 00001fa010_930831.nrm

Page 13: Face Recognition

Implementation Image Lists

Image Lists is a file which lists on each line the same-class normalized images.

Using Perl, we parse the directory list of normalized images and create this file.

This image list is a reference for training to know which images belong to a single class.

SubjectA_Image1.nrm SubjectA_Image2.nrm SubjectA_Image3.nrmSubjectB_Image1.nrm SubjectB_Image2.nrmSubjectC_Image1.nrm SubjectC_Image2.nrm SubjectC_Image3.nrm. . . .

image_list.srt

Page 14: Face Recognition

Implementation What's next?

Need to implement the DF-LDA algorithm for training using normalized files.

Experimentation and Analysis of results.

Page 15: Face Recognition

References

1. Juwei Lu et. al. Face Recognition using LDA-Based Algorithms. IEEE Transactions of Neural Networks. Vol. 14, no. 1, January 2003.

2. Feret Dataset. National Institute of Standards & Technology.

3. Peter Belhumeur et. Al. Eigenfaces vs Fisherfaces: Recognition Using Class Specific Linear Projection. Vol 19. no 7. July 1997.

4. Evaluation of Face Recognition Algorithms. Colorado State University. http://www.cs.colostate.edu/evalfacerec/data/normalization.html