US5774576A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Pattern recognition by unsupervised metric learning
Assignee: NEC Corp
•
Filing: 1995-07-17
•
Granted: 1998-06-30
•
Expires: 2015-07-17
Abstract
Abstract A pattern recognition method uses unsupervised metric learning starting from a mixture of normal densities which explains well observed data. An improved decision rule is provided for selecting the reference database element most likely to correspond to a query.
Inventors
- 👤 Ingemar J. Cox
- 👤 Peter N. Yianilos
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 20:46
CPC Classifications
G; G06; G06F; G06F18/00; G06F18/20; G06F18/22
Claims Analysis
Total Claims: 10
Claim 1
Independent Claim
53 words
The method of pattern recognition comprising the steps of: estimating a metric from a mixture of multi-variate normal densities representing a statistical model of training data; forming a matrix for each element of the mixture; reducing off-diagonal matrix entries to produce an adjusted metric; and using said adjusted metric to classify incoming queries.
Claim 2
Independent Claim
53 words
The method of pattern recognition comprising the steps of: estimating a metric from a mixture of multi-variate normal densities representing a statistical model of training data; forming a matrix for each element of the mixture; reducing off-diagonal matrix entries to produce an adjusted metric; and using said adjusted metric to classify incoming queries.
Claim 3
Dependent Claim
15 words
The method as set forth in claim 1, where said estimating a metric is ##EQU5##
Claim 4
Dependent Claim
14 words
The method as set forth in claim 1, where said training data comprises faces.
Claim 5
Dependent Claim
14 words
The method as set forth in claim 1, where said training data comprises fingerprints.
Claim 6
Independent Claim
57 words
The method of pattern recognition comprising the steps of: estimating a metric from a mathematical form of a mixture of multi-variate normal densities representing a statistical model of training data; forming a matrix for each element of the mixture; reducing off-diagonal matrix entries to produce an adjusted metric; and using said adjusted metric to classify incoming queries.
Claim 7
Independent Claim
57 words
The method of pattern recognition comprising the steps of: estimating a metric from a mathematical form of a mixture of multi-variate normal densities representing a statistical model of training data; forming a matrix for each element of the mixture; reducing off-diagonal matrix entries to produce an adjusted metric; and using said adjusted metric to classify incoming queries.
Claim 8
Dependent Claim
15 words
The method as set forth in claim 5, where said estimating a metric is ##EQU6##
Claim 9
Dependent Claim
14 words
The method as set forth in claim 5, where said training data comprises faces.
Claim 10
Dependent Claim
14 words
The method as set forth in claim 5, where said training data comprises fingerprints.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2001-06-29 | 2001-12-29 | 2002-06-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2005-06-29 | 2005-12-29 | 2006-06-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2009-06-29 | 2009-12-29 | 2010-06-30 | $7700.00 | Window Closed / Lapsed |