FEEDBACK SYSTEMS AND METHODS FOR RECOGNIZING PATTERNS

• US 20090254502A1
• Filed: 02/27/2009
• Published: 10/08/2009
• Est. Priority Date: 02/27/2008
• Status: Active Grant

• Current Assignee: Tsvi Achler
• Original Assignee: Tsvi Achler
• Inventors: Achler, Tsvi
• Granted Patent: US 11,783,566 B2
• Time in Patent Office: Days
• US Class Current: 706/16
• CPC Class Codes: G06F 18/241 relating to the classification model, e.g. parametric or non-parametric approaches
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06V 10/454 Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects

Abstract

Pattern classification systems and methods are disclosed. The pattern classification systems and methods employ one or more classification networks that can parse multiple patterns simultaneously while providing a continuous feedback about its progress. Pre-synaptic inhibition is employed to inhibit feedback connections to permit more flexible processing. Various additional improvements result in highly robust pattern recognition systems and methods that are suitable for use in research, development, and production.

Claims

1. A computer-implemented method for performing pattern classification of objects, comprising:

parsing one or more patterns simultaneously; and

providing progress feedback while parsing.

2. The computer-implemented method of claim 1 further comprising employing pre-synaptic inhibition to inhibit feedback connections.

3. The computer-implemented method of claim 1 further comprising employing a classification network for the performing the pattern classification.

4. A recognition system for performing pattern classification of objects, comprising:

a first module for parsing one or more patterns simultaneously; and

a second module for providing progress feedback while parsing.

5. The recognition system of claim 4 further comprising a third module for inhibiting feedback connections based on pre-synaptic inhibition.

6. The recognition system of claim 4 further comprising a classification network for performing the pattern classification.