Wavelet-based hybrid neurosystem for classifying a signal or an image represented by the signal in a data system
First Claim
1. A system for signal classification comprising:
- means for receiving a series of input signals;
means for transforming said input signals so that characteristics of said signals are represented in the form of wavelet transform coefficients;
means for classifying said signals into at least one distinct category and generating a classification output signal indicative of presence of a distinct category of said at least one category;
means for allowing an operator to specify a basis kernel function for said wavelet transformation; and
means for allowing said operator to specify a selected portion of the coefficients to be processed by the means for classifying and generating an output signal, starting with the largest magnitude of coefficient and proceeding with other coefficients in monotonically descending order of magnitude.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention relates to a system and a method for signal classification. The system comprises a sensor array for receiving a series of input signals such as acoustic signals, pixel-based image signal (such as from infrared images detectors), light signals, temperature signals, etc., a wavelet transform module for transforming the input signals so that characteristics of the signals are represented in the form of wavelet transform coefficients and an array of hybrid neural networks for classifying the signals into multiple distinct categories and generating a classification output signal. The hybrid neural networks each comprise a location neural network for processing data embedded in the frequency versus time location segment of the output of the transform module, a magnitude neural network for processing magnitude information embedded in the magnitude segment of the output of the transform module, and a classification neural network for processing the outputs from the location and magnitude neural networks. A method for processing the signal using the system of the present invention is also described.
56 Citations
9 Claims
-
1. A system for signal classification comprising:
-
means for receiving a series of input signals; means for transforming said input signals so that characteristics of said signals are represented in the form of wavelet transform coefficients; means for classifying said signals into at least one distinct category and generating a classification output signal indicative of presence of a distinct category of said at least one category; means for allowing an operator to specify a basis kernel function for said wavelet transformation; and means for allowing said operator to specify a selected portion of the coefficients to be processed by the means for classifying and generating an output signal, starting with the largest magnitude of coefficient and proceeding with other coefficients in monotonically descending order of magnitude. - View Dependent Claims (2)
-
-
3. A system for signal classification comprising:
-
means for receiving a series of input signals; means for transforming said input signals so that characteristics of said signals are represented in the form of wavelet transform coefficients; means for classifying said signals into at least one distinct category and generating a classification output signal indicative of presence of a distinct category of said at least one category; said classifying means comprising at least one hybrid multi-component neural network system; and each hybrid multi-component neural network system comprising a set of a first location feature, neural network and a second magnitude feature, neural network, and a third classification neural network, the outputs of the set of feature neural networks being parallel fed to the classification neural network. - View Dependent Claims (4, 5, 6, 7, 8, 9)
-
Specification