Date identifying apparatus, date identifying method, and computer-readable recording medium

• US 10,380,416 B2
• Filed: 05/26/2017
• Issued: 08/13/2019
• Est. Priority Date: 02/14/2017
• Status: Active Grant

• Current Assignee: PFU Limited (Fujitsu Limited)
• Original Assignee: PFU Limited (Fujitsu Limited)
• Inventors: Hayashi, Masayoshi
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US15/606,250
• Publication Number:
  US 20180232884A1
• Time in Patent Office: 809 Days
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/768 using context analysis, e.g. recognition aided by known co-occurring patterns
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

A date identifying apparatus includes a segmenting unit that segments date area data, the date area data being data of an area to which a date is estimated to be provided, from image data generated by capturing a medium to which the date is provided, a binarizing unit that executes a binarizing process for binarizing the date area data to generate binarized data, an identification unit that executes an identifying process for executing a recognizing process using deep learning on the binarized data to identify a character string included in the date area data, a correction unit that executes a correcting process according to a kind of a font of the character string on an identified result of the identifying process, and a provision unit that provides, to the image data, date data based on a character string included in the corrected identified result.

Claims

1. A date identifying apparatus comprising:

a memory storing a program; and

a processor coupled to the memory, wherein

when the program is executed by the processor, the processor is configured to;

segment date area data, the date area data being data of an area to which a date is estimated to be provided, from image data generated by capturing a medium to which the date is provided;

execute a binarizing process for binarizing the date area data to generate binarized data;

execute an identifying process for executing a recognizing process using deep learning on the binarized data to identify a character string included in the date area data;

execute a correcting process according to a kind of a font of the character string on an identified result of the identifying process; and

provide, to the image data, date data based on the corrected character string included in the corrected identified result,

when the kind of the font is a Data SEGment (DSEG) font, the processor executes the correcting process based on a number of segments constituting each numeric character included in the identified result, and

when the kind of the font is the font other than the DSEG font, the processor executes the correcting process based on a positional relation between characters included in the identified result.

2. The date identifying apparatus according to claim 1, wherein,

the processor:

respectively executes the binarizing process and the recognizing process for a plurality of times, and

identifies the character string based on the identified result having a largest probability.

3. The date identifying apparatus according to claim 1, wherein,

when identifying in a lump a plurality of characters included in the character string is difficult, the processor individually identifies each of the plurality of characters.

4. A date identifying method comprising:

segmenting date area data, the date area data being data of an area to which a date is estimated to be provided, from image data generated by capturing a medium to which the date is provided;

generating binarized data by binarizing the date area data;

executing an identifying process for executing a recognizing process using deep learning on the binarized data to identify a character string included in the date area data;

correcting an identified result of the identifying process in accordance with a kind of a font of the character string; and

providing, to the image data, date data based on the corrected character string included in the corrected identified result, wherein,

when the kind of the font is a Data SEGment (DSEG) font, correcting the identified result based on a number of segments constituting each numeric character included in the identified result, and

when the kind of the font is the font other than the DSEG font, correcting the identified result based on a positional relation between characters included in the identified result.

5. A non-transitory computer readable recording medium having stored therein a date identifying program, the date identifying program causing a processor to execute a process comprising:

segmenting date area data, the date area data being data of an area to which a date is estimated to be provided, from image data generated by capturing a medium to which the date is provided;

generating binarized data by binarizing the date area data;

executing an identifying process for executing a recognizing process using deep learning on the binarized data to identify a character string included in the date area data;

correcting an identified result of the identifying process in accordance with a kind of a font of the character string; and

providing, to the image data, date data based on the corrected character string included in the corrected identified result, wherein,

when the kind of the font is a Data SEGment (DSEG) font, correcting the identified result based on a number of segments constituting each numeric character included in the identified result, and

when the kind of the font is the font other than the DSEG font, correcting the identified result based on a positional relation between characters included in the identified result.