Variable gain neural network image processing system

Claim 2

A neural-simulating system for processing input stimuli, comprising: a plurality of layers, each layer receiving layer input signals and generating layer output signals, said layer input signals including signals from the input stimuli and ones of said layer output signals from only previous layers within said plurality of layers; each of said plurality of layers including a plurality of neurons operating in parallel on said layer input signals applied to said plurality of layers; and each of said neurons deriving neuron output signals from a continuously differentiable transfer function for each of said neurons based upon a product of a gain weight associated with each of said neurons and a combination of sets of weights associated with said neurons and said layer input signals.

Claim 3

The neural-simulating system of claim 1 and further including: said plurality of neurons within a layer being arranged in groups of neurons operating in parallel on said layer input signals; weight constraining means associated with each of said groups of neurons for causing each of said groups of neurons to extract a different feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals; and gain weight constraining means associated with each of said groups of neurons for causing each of said groups of neurons to extract a different feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals.

Claim 4

The neural-simulating system of claim 2 and further including: adaptive network means associated with each neuron for generating weight correction signals based upon gradient estimate signals and convergence factors signals of each neuron and for processing said weight correction signals to thereby modify said weights associated with each neuron; weight correction signal constraining means associated with each neuron within said groups of neurons for causing each neuron within a group of said neurons to extract the same feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals; adaptive network means associated with each neuron for generating a gain weight correction signal based upon said gradient estimate signals and said convergence factors signals of each neuron and for processing said gain weight correction signals to thereby modify said gain weights associated with each neuron; and gain weight correction signal constraining means associated with each neuron within said groups of neurons for causing each neuron within a group of said neurons to extract the same feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals.

Claim 5

The neural-simulating system of claim 1 and further including: adaptive network means associated with each neuron for generating weight correction signals based upon said gradient estimate signals and said convergence factors signals of each neuron and for processing said weight correction signals to thereby modify said weights associated with each neuron; and adaptive network means associated with each neuron for generating a gain weight correction signal based upon said gradient estimate signals and said convergence factors signals of each neuron and for processing said gain weight correction signals to thereby modify said gain weights associated with each neuron.

Claim 6

A neural-simulating system which is trained on a set of input patterns to the system for subsequent identification of unknown patterns, comprising: clustering means for extracting a training set of patterns representing samples of the input patterns and for updating the training set from the input patterns to the system; a plurality of layers, each layer receiving layer input signals and generating layer output signals, said layer input signals including signals from the training set and ones of said layer output signals from only previous layers within said plurality of layers; each of said plurality of layers including a plurality of neurons operating in parallel on said layer input signals applied to said plurality of layers; each of said neurons deriving neuron output signals from a continuously differentiable transfer function for each of said neurons based upon a product of a gain weight associated with each of said neurons and a combination of sets of weights associated with said neurons and said layer input signals; and adaptive network means associated with each neuron for generating gain weight correction signals based upon gradient estimate signals and convergence factors signals of each neuron and for processing said gain weight correction signals to thereby modify said gain weights associated with each neuron.

Claim 7

A neural-simulating system which is trained on a set of input patterns to the system for subsequent identification of unknown patterns, comprising: clustering means for extracting a training set of patterns representing samples of the input patterns and for updating the training set from the input patterns to the system; a plurality of layers, each layer receiving layer input signals and generating layer output signals, said layer input signals including signals from the training set and ones of said layer output signals from only previous layers within said plurality of layers; each of said plurality of layers including a plurality of neurons operating in parallel on said layer input signals applied to said plurality of layers; each of said neurons deriving neuron output signals from a continuously differentiable transfer function for each of said neurons based upon a product of a gain weight associated with each of said neurons and a combination of sets of weights associated with said neurons and said layer input signals; and adaptive network means associated with each neuron for generating gain weight correction signals based upon gradient estimate signals and convergence factors signals of each neuron and for processing said gain weight correction signals to thereby modify said gain weights associated with each neuron.

Claim 8

The neural-simulating system of claim 5 and further including: said plurality of neurons within a layer being arranged in groups of neurons operating in parallel on said layer input signals; weight constraining means associated with each of said groups of neurons for causing each of said groups of neurons to extract a different feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals; and weight correction signal constraining means associated with each neuron within said groups of neurons for causing each neuron within a group of said neurons to extract the same feature from said layer input signals and for further causing each neuron within a group of neurons to operate on a different portion of said layer input signals.

Claim 9

The neural-simulating system of claim 5 and further including; means for generating neuron output signals from a continuously differentiable transfer function for each of said neurons based upon a product of a gain weight associated with each of said neurons and a combination of sets of weights associated with said neurons and said layer input signals; and adaptive network means associated with each neuron for generating gain weight correction signals based upon gradient estimate signals and convergence factors signals of each neuron and for processing said gain weight correction signals to thereby modify said gain weights associated with each neuron.