Patent US5930392A successfully refreshed from Google Patents.
US5930392A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Classification technique using random decision forests
Assignee: Nokia of America Corp
•
Filing: 1996-07-12
•
Granted: 1999-07-27
•
Expires: 2016-07-12
Abstract
Abstract In a decision-forest classifier in accordance with the invention, a decision forest including multiple decision trees is used to classify "seen" training data and "unseen" data. Each individual tree performs an initial classification based on randomly selected subsets of the data. The classification outcomes by the individual trees are combined using a discriminant process in the decision-forest classier to render the ultimate classification decision.
Inventors
- 👤 Tin Kam Ho
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 18:43
CPC Classifications
G; G06; G06F; G06F18/00; G06F18/20; G06F18/21; G06F18/211; G06F18/24; G06F18/243; G06F18/24323
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
132 words
Apparatus for assigning an image to a selected one of a plurality of classes comprising: a generator for generating a plurality of features based on the values of a plurality of pixels representing said image; a first processor for randomly selecting subsets of features from said plurality of features, and for providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; a plurality of decision-tree classifiers each including a plurality of nonterminal nodes, the nonterminal nodes of each decision-tree classifier, in response to at least one of said sample points, jointly rendering a classification of said image; and a second processor for combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 2
Dependent Claim
132 words
Apparatus for assigning an image to a selected one of a plurality of classes comprising: a generator for generating a plurality of features based on the values of a plurality of pixels representing said image; a first processor for randomly selecting subsets of features from said plurality of features, and for providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; a plurality of decision-tree classifiers each including a plurality of nonterminal nodes, the nonterminal nodes of each decision-tree classifier, in response to at least one of said sample points, jointly rendering a classification of said image; and a second processor for combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 3
Dependent Claim
22 words
The apparatus of claim 4 wherein each feature is a function of the values of selected ones of said plurality of pixels.
Claim 4
Dependent Claim
14 words
The apparatus of claim 2 wherein the selected pixels are neighbors to one another.
Claim 5
Independent Claim
131 words
A method for use in an apparatus comprising a plurality of decision-tree classifiers to assign an image to a selected one of a plurality of classes, each classifier including a plurality of nonterminal nodes, comprising the steps of: generating a plurality of features based on the values of a plurality of pixels representing said image; randomly selecting subsets of features from said plurality of features; providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; jointly rendering by the nonterminal nodes of each decision-tree classifier a classification of said image in response to at least one of said sample points; and combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 6
Independent Claim
131 words
A method for use in an apparatus comprising a plurality of decision-tree classifiers to assign an image to a selected one of a plurality of classes, each classifier including a plurality of nonterminal nodes, comprising the steps of: generating a plurality of features based on the values of a plurality of pixels representing said image; randomly selecting subsets of features from said plurality of features; providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; jointly rendering by the nonterminal nodes of each decision-tree classifier a classification of said image in response to at least one of said sample points; and combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 7
Dependent Claim
22 words
The method of claim 4 wherein each feature is a function of the values of selected ones of said plurality of pixels.
Claim 8
Dependent Claim
14 words
The method of claim 5 wherein the selected pixels are neighbors to one another.
Claim 9
Dependent Claim
127 words
Apparatus for assigning an image to a selected one of a plurality of classes comprising: means for generating a plurality of features based on the values of a plurality of pixels representing said image; means for randomly selecting subsets of features from said plurality of features; means for providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; a plurality of decision-tree classifiers each including a plurality of nonterminal nodes, the nonterminal nodes of each decision-tree classifier, in response to at least one of said sample points, jointly rendering a classification of said image; and means for combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 10
Dependent Claim
127 words
Apparatus for assigning an image to a selected one of a plurality of classes comprising: means for generating a plurality of features based on the values of a plurality of pixels representing said image; means for randomly selecting subsets of features from said plurality of features; means for providing a plurality of sample points, each sample point corresponding to a different one of said subsets of features; a plurality of decision-tree classifiers each including a plurality of nonterminal nodes, the nonterminal nodes of each decision-tree classifier, in response to at least one of said sample points, jointly rendering a classification of said image; and means for combining the rendered classifications from said plurality of decision-tree classifiers to determine said selected class to which said image is assigned.
Claim 11
Dependent Claim
21 words
The apparatus of claim 1 wherein the first processor includes the plurality of decision-tree classifiers, the generator and the second processor.
References Cited (Backward Citations)
US5661820A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-07-26 | 2003-01-25 | 2003-07-27 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-07-26 | 2007-01-25 | 2007-07-27 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-07-26 | 2011-01-25 | 2011-07-27 | $7700.00 | Window Closed / Lapsed |