Information processing apparatus, storage medium, and information processing method
AssigneeCanon Kabushiki Kaisha
InventorMitsuo Kimura
Life of this patent
Estimating the term…
Abstract
A search area is set on a recognition target image, cutout areas are set at a plurality of positions in the search area, images corresponding to the plurality of set cutout areas are extracted, similarities of candidate characters obtained by comparison between the extracted images and dictionary data is weighted in accordance with the positions of the cutout areas. In such a manner, evaluation values of the candidate characters are obtained, and a candidate character with the highest evaluation value among the obtained candidate characters is output as a recognition result. Further, a search area relating to a next character is set based on position information about the cutout area corresponding to the recognition result.
Claims18
An apparatus comprising:
- a first setting unit configured to set a search area on a recognition target image;
- a second setting unit configured to set cutout areas at a plurality of positions in the search area;
- and
- a character detection unit configured to extract images respectively corresponding to the plurality of cutout areas, weight similarities of candidate characters obtained by comparison between the extracted images and dictionary data in accordance with the positions of the cutout areas to obtain evaluation values of the candidate characters, and output, as a recognition result, a candidate character with a highest evaluation value among the obtained candidate characters,
- wherein the first setting unit further sets a search area relating to a next character based on position information of the cutout area corresponding to the recognition result output from the character detection unit.
A non-transitory computer-readable storage medium storing a computer program that causes a computer to perform:
- first setting for setting a search area on a recognition target image;
- second setting for setting cutout areas at a plurality of positions in the search area;
- and
- character detection for extracting images respectively corresponding to the plurality of cutout areas, weighting similarities of candidate characters obtained by comparison between the extracted images and dictionary data in accordance with the positions of the cutout areas to obtain evaluation values of the candidate characters, and outputting, as a recognition result, a candidate character with a highest evaluation value among the obtained candidate characters,
- wherein in the first setting, a search area relating to a next character is set based on position information about the cutout area corresponding to the output recognition result in the character detection.
A method, comprising:
- first setting for setting a search area on a recognition target image;
- second setting for setting cutout areas at a plurality of positions in the search area;
- and
- character detection for extracting images respectively corresponding to the plurality of cutout, weighting similarities of candidate characters obtained by comparison between the extracted images and dictionary data in accordance with the positions of the cutout areas to obtain evaluation values of the candidate characters, and outputting, as a recognition result, a candidate character with a highest evaluation value among the obtained candidate characters,
- wherein in the first setting, a search area relating to a next character is set based on position information about the cutout area corresponding to the recognition result output in the character detection.
Dependencies are inferred from the claim text.