×

Hybrid neural network classifier, systems and methods

  • US 5,943,661 A
  • Filed: 07/11/1991
  • Issued: 08/24/1999
  • Est. Priority Date: 07/11/1991
  • Status: Expired due to Term
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.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×