Information processing method and apparatus
First Claim
Patent Images
1. An information processing method comprising the steps of:
- inputting a plurality of learning patterns;
generating, for each one of the plurality of learning patterns input in said inputting step, one hierarchical structure comprising a plurality of layers based on one learning pattern, the plurality of layers having respective different orders; and
preparing a classification tree comprising a plurality of branches, wherein each of the branches has a plurality of neurons and is developed from an upper layer to a lower layer by selecting at least one neuron of each of the branches, said one neuron having a maximum value for classification efficiency based on the characteristics of each layer of the plurality of hierarchical structures generated for the plurality of learning patterns in said generating step and by developing the selected at least one neuron.
1 Assignment
0 Petitions
Accused Products
Abstract
A classification tree which allows direct recoginition of an input pattern such as image or sound without extra processing such as pre-processing of unprocessed pattern data having high order characteristic variables is prepared. Information processing method and apparatus conduct hierarchical pre-processing for hierarchically pre-processing a learning pattern, prepares a classification tree based on the learning pattern processed by the hierarchical pre-processing and conducts the recognition by using the classification tree.
-
Citations
60 Claims
-
1. An information processing method comprising the steps of:
-
inputting a plurality of learning patterns;
generating, for each one of the plurality of learning patterns input in said inputting step, one hierarchical structure comprising a plurality of layers based on one learning pattern, the plurality of layers having respective different orders; and
preparing a classification tree comprising a plurality of branches, wherein each of the branches has a plurality of neurons and is developed from an upper layer to a lower layer by selecting at least one neuron of each of the branches, said one neuron having a maximum value for classification efficiency based on the characteristics of each layer of the plurality of hierarchical structures generated for the plurality of learning patterns in said generating step and by developing the selected at least one neuron. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 58)
-
-
20. An information processing apparatus comprising:
-
inputting means for inputting a plurality of learning patterns;
generating means for generating, for each one of the plurality of learning patterns input by said inputting means, one hierarchical structure comprising a plurality of layers based on one learning pattern, the plurality of layers having respective different orders; and
classification tree preparation means for preparing a classification tree comprising a plurality of branches wherein each of the branches has a plurality of neurons and is developed from an upper layer to a lower layer by selecting at least one neuron of said each of the branches, said one neuron having a maximum value for classification efficiency based on the characteristics of each layer of the plurality of hierarchical structures generated for the plurality of learning patterns by said generating means, and by developing the selected at least one neuron. - View Dependent Claims (21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 59)
-
-
39. A computer readable medium storing instructions for causing a computer to perform an information processing method, said method comprising the steps of:
-
inputting a plurality of learning patterns;
generating, for each one of the plurality of learning patterns input in said inputting step, one hierarchical structure comprising a plurality of layers based on one learning pattern, the plurality of layers having respective different orders; and
preparing a classification tree comprising a plurality of branches wherein each of the branches has a plurality of neurons and is developed from an upper layer to a lower layer by selecting at least one neuron of said each of the branches, said one neuron having a maximum value for classification efficiency based on the characteristics of each layer of the plurality of hierarchical structures generated for the plurality of learning patterns in said generating step, and by developing the selected at least one neuron. - View Dependent Claims (40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 60)
-
Specification