Pattern identification method, apparatus, and program
First Claim
1. A pattern identification method of identifying a pattern of input data by hierarchically extracting features of the input data, comprising:
- using a processor to perform the steps of;
a first feature extraction step of extracting a feature of a first layer from the input data;
an analysis step of analyzing a distribution of a feature extraction result in the first feature extraction step;
a calculation step of calculating a respective likelihood of extracting from the input data a feature of one of a plurality of categories for features of a second layer, each feature of the second layer corresponding to a combination of features of the first layer, on the basis of the distribution analyzed in the analysis step, wherein the likelihood is calculated by filtering using a mask unique to the feature to be extracted, and integrating results of the filtering;
a selection step of selecting at least one extraction module, among a plurality of extraction modules which extract features of respective categories, whose calculated likelihood of the category for the feature of the second layer to be extracted from the input data is not less than a predetermined value;
a second feature extraction step of causing the selected extraction module to extract a feature of the second layer from the input data; and
a storing step of storing the extracted feature of the second layer in a memory.
1 Assignment
0 Petitions
Accused Products
Abstract
Pattern recognition capable of robust identification for the variance of an input pattern is performed with a low processing cost while the possibility of identification errors is decreased. In a pattern recognition apparatus which identifies the pattern of input data from a data input unit (11) by using a hierarchical feature extraction processor (12) which hierarchically extracts features, an extraction result distribution analyzer (13) analyzes a distribution of at least one feature extraction result obtained by a primary feature extraction processor (121). On the basis of the analytical result, a secondary feature extraction processor (122) performs predetermined secondary feature extraction.
-
Citations
17 Claims
-
1. A pattern identification method of identifying a pattern of input data by hierarchically extracting features of the input data, comprising:
-
using a processor to perform the steps of; a first feature extraction step of extracting a feature of a first layer from the input data; an analysis step of analyzing a distribution of a feature extraction result in the first feature extraction step; a calculation step of calculating a respective likelihood of extracting from the input data a feature of one of a plurality of categories for features of a second layer, each feature of the second layer corresponding to a combination of features of the first layer, on the basis of the distribution analyzed in the analysis step, wherein the likelihood is calculated by filtering using a mask unique to the feature to be extracted, and integrating results of the filtering; a selection step of selecting at least one extraction module, among a plurality of extraction modules which extract features of respective categories, whose calculated likelihood of the category for the feature of the second layer to be extracted from the input data is not less than a predetermined value; a second feature extraction step of causing the selected extraction module to extract a feature of the second layer from the input data; and a storing step of storing the extracted feature of the second layer in a memory. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A pattern identification apparatus for identifying a pattern of input data by hierarchically extracting features of the input data, comprising:
-
first feature extracting means for extracting a feature of a first layer from the input data; analyzing means for analyzing a distribution of a feature extraction result obtained by said first feature extracting means; calculating means for calculating a respective likelihood of extracting from the input data a feature of one of a plurality of categories for features of a second layer, each feature of the second layer corresponding to a combination of features of the first layer on the basis of the distribution analyzed by said analyzing means, wherein the likelihood is calculated by filtering using a mask unique to the feature to be extracted, and integrating results of the filtering; selection means for selecting at least one extraction module, from among a plurality of extraction modules which extract features of respective categories, whose calculated likelihood of the category for the feature of the second layer to be extracted from input data is not less than a predetermined value; second feature extracting means for causing the selected extraction module to extract a feature of the second layer from the input data; and storing means for storing the extracted feature of the second layer in a memory.
-
-
17. A non-transitory computer-readable storage medium on which is stored a pattern identification program for allowing a computer to identify a pattern of input data by hierarchically extracting features of the input data, comprising:
-
a first feature extraction step of extracting a feature of a first layer from the input data; an analysis step of analyzing a distribution of a feature extraction result in the first feature extraction step; a calculation step of calculating a respective likelihood of extracting from the input data a feature of one of a plurality of categories for features of a second layer, each feature of the second layer corresponding to a combination of features of the first layer, on the basis of the distribution analyzed in the analysis step, wherein the likelihood is calculated by filtering using a mask unique to the feature to be extracted, and integrating results of the filtering; a selection step of selecting at least one extraction module, among a plurality of extraction modules which extract features of respective categories, whose calculated likelihood of the category for the feature of the second layer to be extracted from the input data is not less than a predetermined value; a second feature extraction step of causing the selected extraction module to extract a feature of the second layer from the input data; and a storing step of storing the extracted feature of the second layer in a memory.
-
Specification