Back to results
US 2011/0299731 A1Published application
Information processing device and method, and program
AssigneeSony Corporation
InventorsJun Yokono, Kohtaro Sabe
Life of this patent
Estimating the term…
Abstract
An information processing device includes a first calculation unit which calculates a score of each sample image including a positive image in which an object as an identification object is present and a negative image in which the object as the identification object is not present, for each weak identifier of an identifier including a plurality of weak identifiers, a second calculation unit which calculates the number of scores when the negative image is processed, which are scores less than a minimum score among scores when the positive image is processed; and an realignment unit which realigns the weak identifiers in order from a weak identifier in which the number calculated by the second calculation unit is a maximum.
Claims12
An information processing device comprising:
- first calculation means which calculates a score of each sample image including a positive image in which an object as an identification object is present and a negative image in which the object as the identification object is not present, for each weak identifier of an identifier including a plurality of weak identifiers;
- second calculation means which calculates the number of scores when the negative image is processed, which are scores less than a minimum score among scores when the positive image is processed for each weak identifier;
- and
- realignment means which realigns the weak identifiers in order from a weak identifier in which the number calculated by the second calculation means is a maximum.
1 more element
An information processing method comprising the steps of:
- calculating a score of each sample image including a positive image in which an object as an identification object is present and a negative image in which the object as the identification object is not present, for each weak identifier of an identifier including a plurality of weak identifiers;
- calculating the number of scores when the negative image is processed, which are scores less than a minimum score among scores when the positive image is processed for each weak identifier;
- and
- aligning the weak identifiers in order from a weak identifier in which the calculated number is a maximum.
1 more element
A computer-readable program for executing a process comprising the steps of:
- calculating a score of each sample image including a positive image in which an object as an identification object is present and a negative image in which the object as the identification object is not present, for each weak identifier of an identifier including a plurality of weak identifiers;
- calculating the number of scores when the negative image is processed, which are scores less than a minimum score among scores when the positive image is processed for each weak identifier;
- and
- realigning the weak identifiers in order from a weak identifier in which the calculated number is a maximum.
1 more element
An information processing device comprising:
- first calculation means which calculates a score of each sample image, for each weak identifier of an identifier including a plurality of weak identifiers;
- second calculation means which calculates a learning error from the score calculated by the first calculation means;
- and
- realignment means which realigns the weak identifiers in order from a weak identifier in which the learning error calculated by the second calculation means is a minimum.
1 more element
An information processing method comprising the steps of:
- calculating a score of each sample image, for each weak identifier of an identifier including a plurality of weak identifiers;
- calculating a learning error from the calculated score;
- and
- aligning the weak identifiers in order from a weak identifier in which the calculated learning error is a minimum.
1 more element
A computer-readable program for executing a process comprising the steps of:
- calculating a score of each sample image, for each weak identifier of an identifier including a plurality of weak identifiers;
- calculating a learning error from the calculated score;
- and
- realigning the weak identifiers in order from a weak identifier in which the calculated learning error is a minimum.
1 more element
An information processing device comprising:
- calculation means which calculates an average number of weak identifiers when an operation when a sample image is processed is terminated, for each identifier including a plurality of weak identifiers, in which an arrangement of the weak identifiers is different;
- and
- generation means which generates the identifier in which the average number is minimized by performing an operation based on a genetic algorithm and using the average number calculated by the calculation means when the operation is performed.
An information processing method comprising the steps of:
- calculating an average number of weak identifiers when an operation when a sample image is processed is terminated, for each identifier including a plurality of weak identifiers, in which arrangement of the weak identifiers is different;
- and
- generating the identifier in which the average number is minimized by performing an operation based on a genetic algorithm and using the average number when the operation is performed.
A computer-readable program for executing a process comprising the steps of:
- calculating an average number of weak identifiers when an operation when a sample image is processed is terminated, for each identifier including a plurality of weak identifiers, in which arrangement of the weak identifiers is different;
- and
- generating the identifier in which the average number is minimized by performing an operation based on a genetic algorithm and using the average number when the operation is performed.
An information processing device comprising:
- a first calculation unit which calculates a score of each sample image including a positive image in which an object as an identification object is present and a negative image in which the object as the identification object is not present, for each weak identifier of an identifier including a plurality of weak identifiers;
- a second calculation unit which calculates the number of scores when the negative image is processed, which are scores less than a minimum score among scores when the positive image is processed for each weak identifier;
- and
- a realignment unit which realigns the weak identifiers in order from a weak identifier in which the number calculated by the second calculation unit is a maximum.
1 more element
An information processing device comprising:
- first calculation unit which calculates a score of each sample image, for each weak identifier of an identifier including a plurality of weak identifiers;
- second calculation unit which calculates a learning error from the score calculated by the first calculation unit;
- and
- realignment unit which realigns the weak identifiers in order from a weak identifier in which the learning error calculated by the second calculation means is a minimum.
1 more element
An information processing device comprising:
- calculation unit which calculates an average number of weak identifiers when an operation when a sample image is processed is terminated, for each identifier including a plurality of weak identifiers, in which an arrangement of the weak identifiers is different;
- and
- generation unit which generates the identifier in which the average number is minimized by performing an operation based on a genetic algorithm and using the average number calculated by the calculation unit when the operation is performed.
Dependencies are inferred from the claim text.