Image-processing device for document image, image-processing method for document image, and storage medium on which program is stored
An image-processing device includes: a reliability calculation unit configured to calculate reliability of a character recognition result on a document image which is a character recognition target on the basis of a feature amount of a character string of a specific item included in the document image; and an output destination selection unit configured to select an output destination of the character recognition result in accordance with the reliability.
1. An image-processing device comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute the instructions to:
calculate reliability of a result of a character recognition process for each of one or more character strings of a specific item acquired by the character recognition process for a document image which is a target of the character recognition process, on a basis of a feature value relating to a format, which indicates attributes of characters or coordinates of a character string range in the document image, of the specific item, which is specified in advance among a plurality of items indicating kinds of information included in the document image, and calculate reliability of the result of the character recognition process for the document image on a basis of the reliability of each of the one or more character strings;
compare the reliability of the document image with a threshold value; and
select an output destination of the result of the character recognition process for the document image in accordance with a comparison result between the reliability of the document image and the threshold value.
2. The image-processing device according to claim 1 ,
wherein the at least one processor is configured to execute the instructions to:
calculate the reliability of each of the one or more character strings of the specific item acquired by the character recognition process, on a basis of feature values, which are recorded in advance based on learning results obtained using a plurality of document images, of character strings of the specific item for each specific item of the plurality of document images.
3. The image-processing device according to claim 2 ,
wherein the at least one processor is configured to execute the instructions to:
calculate the reliability of each of the one or more character strings, on a basis of a degree of variation of the feature value in the feature values recorded in advance based on the learning results obtained using the plurality of document images.
4. The image-processing device according to claim 2 ,
wherein the at least one processor is configured to execute the instructions to:
calculate the reliability of each of the one or more character strings, on a basis of a degree of deviation of the feature value in the document image which is the target of the character recognition process with respect to the feature values recorded in advance based on the learning results obtained using the plurality of document images.
5. An image-processing method, comprising:
calculating reliability of a result of a character recognition process for each of one or more character strings of a specific item acquired by the character recognition process for a document image which is a target of the character recognition process, on a basis of a feature value relating to a format, which indicates attributes of characters or coordinates of a character string range in the document image, of the specific item, which is specified in advance among a plurality of items indicating kinds of information included in the document image, and calculating reliability of the result of the character recognition process for the document image on a basis of the reliability of each of the one or more character strings;
comparing the reliability of the document image with a threshold value; and
selecting an output destination of the result of the character recognition process for the document image in accordance with a comparison result between the reliability of the document image and the threshold value.
6. A non-transitory computer-readable storage medium that stores a program causing a computer to perform processes of:
calculating reliability of a result of a character recognition process for each of one or more character strings of a specific item acquired by the character recognition process for a document image which is a target of the character recognition process, on a basis of a feature value relating to a format, which indicates attributes of characters or coordinates of a character string range in the document image, of the specific item, which is specified in advance among a plurality of items indicating kinds of information included in the document image, and calculating reliability of the result of the character recognition process for the document image on a basis of the reliability of each of the one or more character strings;
comparing the reliability of the document image with a threshold value; and
selecting an output destination of the result of the character recognition process for the document image in accordance with a comparison result between the reliability of the document image and the threshold value.