System and method of face recognition using proportions of learned model

1. A method for classifying facial image data, the method comprising the steps of:

a) training a neural network classifier device for recognizing one or more facial images and

obtaining corresponding learned models of the facial images used for training;

b) inputting a vector including data representing a portion of an unknown facial image to be recognized into said classifier;

c) classifying said portion of said unknown facial image according to a classification method at each iteration,

comparing a portion of the unknown image against a corresponding portion of the learned model image for each class, and

obtaining a confidence score for each classified portion;

d) repeating step b) and c) using a different portion of said unknown facial image at each iteration; and,

e) identifying a single class result from said different portions input to said classifier, applying a rule to said confidence scores to obtain said single class result, said confidence score is a probability measure that a current portion of an unknown facial image is identified with a class, said applied rule including obtaining class having majority of class labels determined for each unknown facial image.

2. The method of claim 1 , wherein said classifying step c) includes decreasing at each iteration, the portion of the unknown image being tested and,

comparing the decreased portion of the unknown image against a corresponding decreased portion of the learned model image for each class.

3. The method of claim 2 , wherein said portions are decreased from 100% of the unknown facial image to 50% of the unknown facial image at equal decrements.

4. The method of claim 1 , wherein a Radial Basis Function Network is implemented for training and classifying each image portion.

5. The method of claim 1 , wherein the classifying step c) comprises outputting a class label identifying a class to which the detected unknown facial image portion corresponds to and a probability value indicating the probability with which the unknown facial image pattern belongs to the class.

6. A method for classifying facial image data, the method comprising the steps of:

a) training a neural network classifier device for recognizing one or more facial images and

obtaining corresponding learned models of the facial images used for training, wherein a Radial Basis Function Network is implemented for training and classifying each image portion, said training step comprises:

i) initiating the Radial Basis Function Network, the initializing step comprising the steps of:

fixing the network structure by selecting a number of basis functions F, where each basis function I has the output of a Gaussian non-linearity,

determining the basis function means μ I where I=1, . . . , F, using a K-means clustering algorithm,

determining the basis function variances σ I 2 , and

determining a global proportionality factor H, for the basis function variances by empirical search;

ii) presenting the training, the presenting step comprising the steps of:

inputting training patterns X(p) and their class labels C(p) to the classification method, where the pattern index is p=1, . . . , N,

computing the output of the basis function nodes y I (p), F, resulting from pattern X(p);

computing the F×F correlation matrix R of the basis function outputs; and

computing the F×M output matrix B, where d j is the desired output and M is the number of output classes and j=1, . . . , M, and

iii) determining weights, the determining step comprising the steps of:

inverting the F×F correlation matrix R to get R − ; and

solving for the weights in the network;

b) inputting a vector including data representing a portion of an unknown facial image to be recognized into said classifier;

c) classifying said portion of said unknown facial image according to a classification method;

d) repeating step b) and c) using a different portion of said unknown facial image at each iteration; and,

e) identifying a single class result from said different portions input to said classifier.

7. A method for classifying facial image data, the method comprising the steps of:

a) training a neural network classifier device for recognizing one or more facial images and

obtaining corresponding learned models of the facial images used for training, wherein a Radial Basis Function Network is implemented for training and classifying each image portion, wherein said training step comprises:

i) initiating the Radial Basis Function Network, the initializing step comprising the steps of:

fixing the network structure by selecting a number of basis functions F, where each basis function I has the output of a Gaussian non-linearity;

determining the basis function means μ I where I=1, . . . , F, using a K-means clustering algorithm;

determining the basis function variances σ I 2 ; and

determining a global proportionality factor H, for the basis function variances by empirical search;

ii) presenting the training, the presenting step comprising the steps of:

inputting training patterns X(p) and their class labels C(p) to the classification method, where the pattern index is p=1, . . . , N,

computing the output of the basis function nodes y I (p), F, resulting from pattern X(p),

computing the F×F correlation matrix R of the basis function outputs, and

computing the F×M output matrix B, where d j is the desired output and M is the number of output classes and j=1, . . . , M and

iii) determining weights, the determining step comprising the steps of:

inverting the F×F correlation matrix R to get R −1 ; and

solving for the weights in the network;

b) inputting a vector including data representing a portion of an unknown facial image to be recognized into said classifier;

c) classifying said portion of said unknown facial image according to a classification method, the classifying step further comprising:

presenting each X test portion at each iteration to the classification method and classifying each X test by computing the basis function outputs, for all F basis functions, computing output node activations, and selecting the output Z j with the largest value and classifying the X test portion as a class j;

d) repeating step b) and c) using a different portion of said unknown facial image at each iteration; and,

e) identifying a single class result from said different portions input to said classifier.

8. An apparatus for classifying facial image data comprising:

a neural network classifier device trained for recognizing one or more facial images and generating corresponding learned models associated with the facial images used for training;

means for iteratively inputting a vector each including data representing a portion of an unknown facial image to be recognized into said classifier, a different image portion being input to said classifier at each iteration, said classifier device classifying each said portion of said unknown facial image according to a classification method;

means for identifying a single class result from said different portions input to said classifier.

9. The apparatus of claim 8 , wherein said classifier includes:

a mechanism for comparing a portion of the unknown image against a corresponding portion of the learned model image for each class, at each iteration; and, obtaining a confidence score for each classified portion.

10. The apparatus of claim 9 , wherein said means for identifying applies a rule to said confidence scores to obtain said single class result.

11. The apparatus of claim 9 , including mechanism for decreasing each portions of each unknown facial image being tested at each iteration and, comparing the decreased portion of the unknown image against a corresponding decreased portion of the learned model image for each class.

12. The apparatus of claim 11 , wherein said portions are decreased from 100% of the unknown facial image to 50% of the unknown facial image at equal decrements.

13. The apparatus of claim 8 , wherein a Radial Basis Function Network is implemented for training and classifying each image portion.

14. An apparatus for classifying facial image data comprising:

a neural network classifier device trained for recognizing one or ore facial images and generating corresponding learned models associated with the facial images used for training;

means for iteratively inputting a vector each including data representing a portion of an unknown facial image to be recognized into said classifier, a different image portion being input to said classifier at each iteration, said classifier device classifying each said portion of said unknown facial image according to a classification method,

said classifier includes a mechanism for comparing a portion of the unknown image against a corresponding portion of the learned model image for each class, at each iteration and obtaining a confidence score for each classified portion,

said confidence score is a probability measure that a current portion of an unknown facial image is identified with a class, said applied rule including identifying class having majority of class labels determined for each unknown facial image; and

means for identifying a single class result from said different portions input to said classifier, said means for identifying applies a rule to said confidence scores to obtain said single class result.

15. A computer-readable medium embodying a program of instructions to perform method steps for classifying facial image data, the method comprising the steps of:

a) training a neural network classifier device for recognizing one or more facial images and

obtaining corresponding learned models the facial images used for training;

b) inputting a vector including data representing a portion of an unknown facial image to be recognized into said classifier;

c) classifying said portion of said unknown facial image according to a classification method at each iteration,

comparing of the unknown image against a corresponding portion of the learned model image for each class, and

obtaining a confidence score for each classified portion;

d) repeating step b) and c) using a different portion of said unknown facial image at each iteration; and,

e) identifying a single class result from said different portions input to said classifier, applying a rule to said confidence score to obtain said single class result, said confidence score is a probability measure that a current portion of an unknown facial image is identified with a class, said applied rule including obtaining class having majority of class labels determined for each unknown facial image.

16. A method for classifying facial image data, the method comprising:

training a classifier device for recognizing one or more facial images and obtaining corresponding learned models the facial images used for training;

inputting a vector including data representing a portion of an unknown facial image to be recognized into said classifier;

classifying said portion of said unknown facial image according to a classification method;

repeating the inputting and classifying using a different portion of said unknown facial image at each iteration; and,

identifying a single class result from said different portions input to said classifier; and wherein:

the classifying includes: at each iteration, comparing a portion of the unknown image against a corresponding portion of the learned model image for each class; and obtaining a confidence score for each classified portion, the confidence score being a probability measure that a current portion of an unknown facial image is identified with a class, an applied rule including obtaining class having majority of class labels determined for each unknown facial image; and

the identifying includes applying the rule to said confidence scores to obtain said single class result.

17. A method for classifying facial image data, the method comprising:

training a classifier device for recognizing one or more facial images and obtaining corresponding learned models the facial images used for training;

inputting a vector including data representing a portion of an unknown facial image to be recognized into the classifier;

classifying the portion of the unknown facial image according to a classification method;

repeating the inputting and classifying using a different portion of the unknown facial image at each iteration; and,

identifying a single class result from the different portions input to the classifier;

and wherein:

a Radial Basis Function Network is implemented for training and classifying each image portion; and

the training includes:

initiating the Radial Basis Function Network, the initializing including: fixing the network structure by selecting a number of basis functions F, where each basis function I has the output of a Gaussian non-linearity; determining the basis function means μ I where I=1, . . . , F, using a K-means clustering algorithm; determining the basis function variances σ I 2 ; and determining a global proportionality factor H, for the basis function variances by empirical search;

presenting the training, the presenting including: inputting training patterns X(p) and their class labels C(p) to the classification method, where the pattern index is p=1, . . . , N; computing the output of the basis function nodes Y I (p), F, resulting from pattern X(p); computing the F×F correlation matrix R of the basis function outputs; and computing the F×M output matrix B, where d j is the desired output and M is the number of output classes and j=1, . . . , M; and

determining weights, the determining including: inverting the F×F correlation matrix R to get R −1 ; and solving for the weights in the network.

18. The method of claim 17 , wherein the classifying includes:

presenting each Xtest portion at each iteration to the classification method; and

classifying each Xtest by:

computing the basis function outputs, for all F basis functions;

computing output node activations; and

selecting the output zj with the largest value and classifying the Xtest portion as a class j.

19. Apparatus for classifying facial image data comprising:

a classifier device trained for recognizing one or ore facial images and generating corresponding learned models associated with the facial images used for training;

means for iteratively inputting a vector each including data representing a portion of an unknown facial image to be recognized into the classifier, a different image portion being input to the classifier at each iteration, the classifier device classifying each the portion of the unknown facial image according to a classification method;

means for identifying a single class result from the different portions input to the classifier,

and wherein:

the classifier includes: a mechanism for comparing a portion of the unknown image against a corresponding portion of the learned model image for each class, at each iteration; and, obtaining a confidence score for each classified portion;

the means for identifying applies a rule to the confidence scores to obtain the single class result; and

the confidence score is a probability measure that a current portion of an unknown facial image is identified with a class, the applied rule including identifying class having majority of class labels determined for each unknown facial image.