Providing in-line previews of a source image for aid in correcting OCR errors

• US 9,501,853 B2
• Filed: 01/09/2015
• Issued: 11/22/2016
• Est. Priority Date: 01/09/2015
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Khan, Sayeed Ullah, Namata, Subrato, Dwivedi, Uttam
• Primary Examiner(s): Williams, Kimberly A
• Application Number:
  US14/593,431
• Publication Number:
  US 20160203625A1
• Time in Patent Office: 683 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 3/033 Pointing devices displaced or positioned by the user , e.g. mice, trackballs, pens or joysticks; Accessories therefor digitisers characterised by the transducing means G06F3/041
• CPC Class Codes: G06F 3/0485 Scrolling or panning
• CPC Class Codes: G06F 40/117 Tagging; Marking up details of markup languages G06F40/143; Designating a block; Setting of attributes style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1456 based on user interactions
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks

Abstract

The present disclosure is directed toward systems and methods for assisting users in correcting OCR errors. For example, systems and methods described herein involve identifying the position of a cursor within a machine-readable document. Systems and methods described herein also involve identifying corresponding position co-ordinates in a source image, as well as, capturing an image preview from the source image based on the corresponding position co-ordinates. Systems and methods described herein may also involve providing the preview of the source image within the machine-readable document.

Claims

1. In a digital medium environment for editing machine-readable documents generated from non-machine-readable documents, a method for aiding in correcting a machine-readable document, comprising:

converting, by at least one processor, an image having text into a document with machine-readable text;

detecting, by the at least one processor, a position of a cursor within the document with machine-readable text;

determining, by the at least one processor, a position in the image corresponding to the detected position of the cursor within the document with machine-readable text;

capturing, upon determining the position in the image and by the at least one processor, a portion of the image proximate the position in the image corresponding to the detected position of the cursor within the document with machine-readable text; and

overlaying, by the at least one processor, the captured portion of the image on the document with machine-readable text adjacent to the detected position of the cursor.

2. The method of claim 1, wherein detecting the position of the cursor within the document with machine-readable text comprises identifying co-ordinates of the cursor with respect to the document with machine-readable text.

3. The method of claim 2, wherein determining the position in the image corresponding to the detected position of the cursor within the document with machine-readable text comprises identifying a co-ordinate position in the image.

4. The method of claim 3, wherein capturing the portion of the image proximate the position in the image corresponding to the detected position of the cursor within the document with machine-readable text comprises positioning an image capture box about the co-ordinate position in the image and capturing the portion of the image within the image capture box.

5. The method of claim 4, further comprising adjusting a parameter of the image capture box based on a characteristic of the image proximate to the co-ordinate position in the image.

6. The method of claim 5, wherein adjusting a parameter of the image capture box based on a characteristic of the image proximate to the co-ordinate position in the image comprises:

determining that the image capture box overlaps a content boundary;

shifting the image capture box away from the boundary; and

wherein overlaying the captured portion of the image on the document with machine-readable text comprises overlaying the captured portion of the image on the document with machine-readable text adjacent to the position of the cursor and offset by the content boundary.

7. The method of claim 6, wherein the content boundary is at least one of a side margin, a start of a paragraph, an illustration, or an end of a paragraph.

8. The method of claim 1, further comprising determining that the detected position of the cursor is near a user interface window boundary, and wherein overlaying the captured portion of the image comprises overlaying the captured portion of the image away from the user interface window boundary.

9. The method of claim 1, further comprising generating one or more bounding boxes in connection with the document having machine-readable text that surround one or more areas of content in the document having machine-readable text.

10. The method of claim 9, wherein an area of content comprises a word, line, a paragraph of text, or an image.

11. The method of claim 1, further comprising:

identifying one or more potential optical character recognition (“

OCR”

) errors within the document with machine-readable text, the one or more potential OCR errors corresponding to converted text having a confidence value below an OCR error threshold level;

receiving a cursor input from a user that locates the position of the cursor within the document with machine-readable text over a potential OCR error of the one or more potential OCR errors; and

wherein detecting the position of the cursor within the document within the machine-readable text comprises determining that the position of the cursor input is over the potential OCR error.

12. The method of claim 11, further comprising:

receiving, from a user, a selection of an option to edit the document with machine-readable text; and

wherein identifying the one or more potential OCR errors within the document with machine-readable text, detecting the position of the cursor within the document with machine-readable text, determining the position in the image corresponding to the detected position of the cursor within the document with machine-readable text, capturing the portion of the image proximate the position in the image corresponding to the detected position of the cursor within the document with machine-readable text, overlaying the captured portion of the image on the document with machine-readable text adjacent to the detected position of the cursor are preformed based on receiving the selection of the option to edit the document.

13. The method of claim 12, further comprising:

receiving, by the user, text input correcting the potential OCR error within the document with machine-readable text, the corrected potential OCR error being located at a position in the document with machine-readable text shown in the captured portion of the image; and

resizing the captured portion of the image based on the user correcting the potential OCR error.

14. In a digital medium environment for machine-readable documents generated from non-machine-readable documents, a method for aiding in correcting a machine-readable document, comprising:

identifying, by at least one processor, a first position of a cursor within an optical character recognized document generated from an image having recognizable text;

obtaining, by the at least one processor, a first portion of the image having recognizable text corresponding to the first position of the cursor within the optical character recognized document;

providing, by the at least one processor, a preview of the image, within the optical character recognized document adjacent to the first position of the cursor, the preview including the first portion of the image having recognizable text;

detecting, by the at least one processor, movement of the cursor within the optical character recognized document from the first position to a second position;

obtaining, by the at least one processor, a second portion of the image having recognizable text corresponding to the second position of the cursor within the optical character recognized document; and

based on the detected movement in the position of the cursor;

updating, by the at least one processor, the preview of the image to include the second portion of the image having recognizable text; and

moving the preview of the image from adjacent the first position to adjacent the second position of the cursor.

15. The method as recited in claim 14, wherein providing, by the at least one processor, the preview of the image, within the optical character recognized document adjacent to the first position of the cursor comprises providing an image preview box containing the preview of the image within the optical character recognized document.

16. The method as recited in claim 15, wherein moving the preview of the image from adjacent the first position to adjacent the second position of the cursor comprises dynamically scrolling the image preview box from the adjacent the first position to adjacent the second position.

17. The method as recited in claim 16, further comprising dynamically updating the preview of the image within the image preview box while dynamically scrolling the image preview box.

18. The method as recited in claim 17, further comprising:

detecting a speed of the movement of the cursor; and

scrolling the image preview box at the detected speed.

19. A system for aiding in correcting a machine-readable document, the system comprising:

at least one processor; and

at least one non-transitory computer readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the system to;

convert an image having text into a document with machine-readable text;

detect a position of a cursor within the document with machine-readable text;

determine a position in the image corresponding to the detected position of the cursor within the document with machine-readable text;

capture, upon determining the position in the image, a portion of the image proximate the position in the image corresponding to the detected position of the cursor within the document with machine-readable text; and

overlay the captured portion of the image on the document with machine-readable text adjacent to the detected position of the cursor.

20. The system as recited in claim 19, further comprising instructions thereon that, when executed by the at least one processor, cause the system to:

provide the captured portion of the image within an image preview box;

detect movement of the cursor within the document with machine-readable text;

dynamically move the image preview box within the document with machine-readable text such that the image preview box stays proximate the cursor; and

dynamically update the portion of the image within the image preview box while moving image preview box within the document with machine-readable text such that the portion of the image displayed within the image preview box corresponds with the position of the curser as the curser and the image preview box move.