Pattern recognition device for performing classification using a candidate table and method thereof
AssigneeFujitsu Limited
InventorsKATSUHITO FUJIMOTO, HIROSHI KAMADA
Life of this patent
Estimating the term…
Abstract
The value of a reference feature vector calculated from a feature vector of an input pattern is transformed into a candidate category set, by a mapping described in a candidate table. Then, pattern recognition is performed using the candidate category set. By suitably setting the mapping, a high-speed process is performed while maintaining a recognition accuracy.
Claims31
A pattern recognition device, comprising:
- table storing means for storing a candidate table describing information required for generating a mapping whose input is a value of a reference feature vector calculated from a feature vector of a pattern, and whose output is a candidate category set;
- and
- candidate category calculating means for obtaining a candidate category set corresponding to a value of a given reference feature vector by using the candidate table, and outputting the obtained candidate category set.
A pattern recognition device, comprising:
- a plurality of table storing means for respectively storing a candidate table describing information required for generating a mapping whose input is a value of a reference feature vector calculated from a feature vector of a pattern, and whose output is a candidate category set;
- a plurality of candidate category calculating means, which are respectively arranged for said plurality of table storing means, for respectively obtaining a candidate category set corresponding to a value of a given reference feature vector by using the candidate table, and outputting the obtained candidate category set;
- and
- category screening means for screening a plurality of candidate category sets output from said plurality of candidate category calculating means, and outputting a result of screening.
A pattern recognition device, comprising:
- storing means for storing a correspondence between feature amount data indicating a feature of a pattern and a candidate category set;
- and
- candidate category calculating means for obtaining a candidate category set corresponding to given feature amount data by using the correspondence, and outputting the obtained candidate category set.
A computer-readable storage medium, when used by a computer, for making the computer perform the functions of:
- obtaining a candidate category set corresponding to a value of a given reference feature vector by using a candidate table describing information required for generating a mapping whose input is a value of a reference feature vector calculated from a feature vector of a pattern, and whose output is a candidate category set;
- and
- outputting the obtained candidate category set.
A computer-readable storage medium, when used by a computer, for making the computer perform the functions of:
- obtaining a candidate category set corresponding to given feature amount data by using a correspondence between feature amount data indicating a feature of a pattern and a candidate category set;
- and
- outputting the obtained candidate category set.
A pattern recognition method, comprising the steps of:
- obtaining a candidate category set corresponding to a value of a given reference feature vector by using a candidate table describing information required for generating a mapping whose input is the value of the reference feature vector calculated from a feature vector of a pattern, and whose output is a candidate category;
- and
- outputting the obtained candidate category set.
A pattern recognition method, comprising the steps of:
- obtaining a candidate category set corresponding to given feature amount data by using a correspondence between feature amount data indicating a feature of a pattern and a candidate category set;
- and
- outputting the obtained candidate category set.
Dependencies are inferred from the claim text.