Image processing apparatus, method, and medium for character recognition

1. An image processing apparatus comprising:

one or more processors; and

memory storing executable instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

accepting character information about a character image in a character region in an image;

performing character recognition on the character image in the character region to obtain a plurality of character recognition results, wherein the performing comprises:

separating the character image into at least one single-character candidate region corresponding to one character in the character image, wherein the character region is set by a pointing device, and

performing the character recognition on the separated at least one single-character candidate region and outputting the plurality of character recognition results for the at least one single-character candidate region;

selecting a character recognition result, from among the plurality of character recognition results, that matches the accepted character information, wherein the selecting comprises:

checking each of the plurality of character recognition results for the at least one single-character candidate region against text information in the accepted character information, the text information corresponding to a character in the at least one single-character candidate region, and

selecting a character recognition result, from among the plurality of character recognition results, that matches the text information;

associating the selected character recognition result with the character image to generate an output result, wherein the output result is a combination of the character image and a character code; and

outputting the output result as training data for subsequent character recognition processing.

2. The image processing apparatus according to claim 1 ,

wherein the separating comprises separating the character image into a plurality of single-character candidate regions.

3. The image processing apparatus according to claim 1 , wherein the memory stores further executable instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

setting the character region including the character image, for the image,

wherein the performing comprises processing the set character region.

4. The image processing apparatus according to claim 1 , wherein the memory stores further executable instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

generating the at least one single-character candidate region from one or more character components included in the character image,

wherein the separating comprises separating the character image into the generated at least one single-character candidate region, as a target for character recognition.

5. An image processing apparatus comprising:

one or more processors; and

memory storing executable instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

accepting character information about a character image in a character region in an image;

performing character recognition on the character image in the character region to obtain a plurality of character recognition results, wherein the performing comprises:

separating the character image into at least one single-character candidate region corresponding to one character in the character image, wherein the character region is set by a pointing device, and

performing the character recognition on the separated at least one single-character candidate region and outputting the plurality of character recognition results for the at least one single-character candidate region;

generating the at least one single-character candidate region from one or more character components included in the character image, wherein the separating comprises separating the character image into the generated at least one single-character candidate region, as a target for character recognition;

selecting a character recognition result, from among the plurality of character recognition results, that matches the accepted character information;

associating the selected character recognition result with the character image to generate an output result, wherein the output result is a combination of the character image and a character code; and

outputting the output result as training data for subsequent character recognition processing,

wherein, in the case where none of the plurality of character recognition results match the accepted character information, the generating comprises generating a new single-character candidate region.

6. An image processing apparatus comprising:

one or more processors; and

memory storing executable instructions which, when executed by the one or more processors, cause the one or more processors to perform the steps of:

accepting character information about a character image in a character region in an image;

performing character recognition on the character image in the character region to obtain a plurality of character recognition results, wherein the performing comprises:

separating the character image into at least one single-character candidate region corresponding to one character in the character image, wherein the character region is set by a pointing device, and

performing the character recognition on the separated at least one single-character candidate region and outputting the plurality of character recognition results for the at least one single-character candidate region;

generating the at least one single-character candidate region from one or more character components included in the character image, wherein the separating comprises separating the character image into the generated at least one single-character candidate region, as a target for character recognition;

selecting a character recognition result, from among the plurality of character recognition results, that matches the accepted character information;

associating the selected character recognition result with the character image to generate an output result, wherein the output result is a combination of the character image and a character code; and

outputting the output result as training data for subsequent character recognition processing,

wherein the generating comprises generating different single-character candidate regions by adding character components, one by one, wherein the number of character components to be added is a predetermined number.

7. The image processing apparatus according to claim 6 ,

wherein, in the case where none of the plurality of character recognition results match the accepted character information and where the number of character components in the latest generated different single-character candidate region is equal to or more than the predetermined number, the generating comprises issuing a notification that no matched character recognition results are present.

8. The image processing apparatus of claim 1 , wherein the accepting comprises accepting the character code, representing the character image, as the character information.

9. The image processing apparatus of claim 1 , wherein the selecting comprises selecting the character recognition result by comparing the accepted character information, which has not undergone character recognition processing, with the plurality of character recognition results to select the character recognition result which matches the accepted character information.