Hybrid neural network classifier, systems and methods
First Claim
Patent Images
1. A method of training a neural network classifier, comprising the steps of:
- (a) providing a first set of target points Z1, Z2, . . . ZL in a feature space;
(b) forming an estimated target probability density P on said feature space from said target points Z1, Z2, . . . ZL ;
(c) providing a second set of target points W1, W2, . . . WM in said feature space;
(d) defining a threshold T from the number of Wj with P(Wj)>
T and the number of Wj with P(Wj)<
T;
(e) providing a third set of points X1, X2, . . . XN in said feature space, and forming a set of pairs (Xj, Yj) where Yj is "target" when P(Xj)>
T and Yj is "clutter" when P(Xj)<
T; and
(f) using the pairs (X1, Y1), (X2, Y2), . . . , (Xj, Yj), . . . , (XN, YN) as input/output pairs to train a neural network classifier.
2 Assignments
0 Petitions
Accused Products
Abstract
A method of and system for parallelizing an program, comprising the steps of inputting an algorithm, operating said algorithm on selected data inputs to generate representative outputs, inputting representative outputs into parallelizing algorithms, and outputting a parallel implementation of said algorithm. In particular, this provides a parallel framework for target classification and pattern recognition procedures.
-
Citations
4 Claims
-
1. A method of training a neural network classifier, comprising the steps of:
-
(a) providing a first set of target points Z1, Z2, . . . ZL in a feature space; (b) forming an estimated target probability density P on said feature space from said target points Z1, Z2, . . . ZL ; (c) providing a second set of target points W1, W2, . . . WM in said feature space; (d) defining a threshold T from the number of Wj with P(Wj)>
T and the number of Wj with P(Wj)<
T;(e) providing a third set of points X1, X2, . . . XN in said feature space, and forming a set of pairs (Xj, Yj) where Yj is "target" when P(Xj)>
T and Yj is "clutter" when P(Xj)<
T; and(f) using the pairs (X1, Y1), (X2, Y2), . . . , (Xj, Yj), . . . , (XN, YN) as input/output pairs to train a neural network classifier. - View Dependent Claims (2, 3, 4)
-
Specification