OCR image preprocessing method for image enhancement of scanned documents

• US 5,594,815 A
• Filed: 05/22/1995
• Issued: 01/14/1997
• Est. Priority Date: 10/19/1992
• Status: Expired due to Term

• Current Assignee: Accusoft Corporation
• Original Assignee: TMSSequoia Co.
• Inventors: Allen, Dana R., Fast, Bruce B.
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): Tran, Phuoc
• Application Number:
  US08/447,215
• Time in Patent Office: 603 Days
• Field of Search: 382/254, 382/274, 382/275, 382/289, 382/290, 382/291, 382/294, 382/296, 382/245, 358/261.1, 358/462, 358/463
• US Class Current: 382/254
• CPC Class Codes: G06T 9/005 Statistical coding, e.g. Huffman, run length coding
• CPC Class Codes: G06V 10/20 Image preprocessing
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1478 of characters or characters lines
• CPC Class Codes: G06V 30/155 Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines
• CPC Class Codes: G06V 30/164 Noise filtering
• CPC Class Codes: H04N 1/38 Circuits or arrangements for blanking or otherwise eliminating unwanted parts of pictures H04N1/387 takes precedence
• CPC Class Codes: H04N 1/3878 Skew detection or correction
• CPC Class Codes: H04N 1/409 Edge or detail enhancement; Noise or error suppression

Abstract

The process for enhancing images of scanned documents identifies a variety of items in the scanned document which could make optical character recognition and other document image processing difficult or impossible. These items include identifying skew, registration, speck, lines intersecting printed matter, reverse printing, shaded printed matter. The process allows to register such items, take steps to correct skew and image registration, as well as reverse invert printing. Items to be removed are listed for deletion in one step. The image of the scanned document is stored and processed as a run-lngth coded image, except for a some operations in which parts of the run length coded image are converted to pixel image code for particular substeps of the process. The result of such a substep, e.g. a modified image, is then re-converted to run-length code.

Claims

1. A process for enhancing an image of a scanned document having boundaries, said image including a plurality of lines of textual matter, said plurality of lines potentially having at least one skew relative to the boundaries of said image, requiring registration, containing inverted text matter, undesired dots, and vertical and horizontal lines, said process comprising

a step for providing a primary scan line run-length coded image of said document;

a step for determining said skew;

a step for reducing said skew;

a step for determining said registration;

a step for correcting said registration;

a step for detecting said inverted text matter;

a step for reversing said inverted text matter;

a step for detecting said undesired dots;

a step for deleting said undesired dots;

a step for detecting said vertical lines;

a step for eliminating said vertical lines;

a step for detecting said horizontal lines;

a step for eliminating said horizontal lines.

2. A process for enhancing images of scanned documents as claimed in claim 1, wherein said process steps for deleting said undesired dots, vertical and horizontal lines eliminate run-length coded segments by adding the segment length of the segment to be deleted to the sum of the length of the adjacent run-length segments.

3. A process for enhancing images of scanned documents as claimed in claim 1, wherein said inverted text matter has foreground run-lengths representing a first color and background run-lengths representing a second color, wherein said first color is lighter than said second color, wherein a non-inverted text matter has foreground run-lengths representing a third color and background run-lengths representing a fourth color, wherein said fourth color is lighter than said third color, and

said process steps for detecting and reversing said inverted text matter reversed said inverted text matter by combining adjacent segment lengths of said Second color and said fourth color at the border of said inverted text matter to form one segment length of said second color, thereby replacing run-lengths of said first color with run-length of said second color.

4. A process for enhancing images of scanned documents as claimed in claim 1, wherein said process includes a step for converting said primary run-length coded image into a secondary run-length coded image having run-length segments normal to the run-length segments of said primary run-length coded image.

5. A process for enhancing images of scanned documents as claimed in claim 1, wherein said process includes

a step for selectively converting a section of said primary run-length coded image into a pixel map,

applying a process substep to said selected section, and

a step for reconverting said so processed pixel map into a primary run-length coded image.

6. A process for enhancing images of scanned documents as claimed in claim 1, wherein said image is run-length coded for horizontal scan lines, said process further includes a step for converting said image to a run-length coded image with vertical scan lines.

7. A process for enhancing images of scanned documents as claimed in claim 1, wherein said image is run-length coded for horizontal scan lines, said process further includes a step for converting said image to a pixel mapped image.

8. A process for enhancing an image of a scanned document having boundaries, said image including a plurality of lines of textual matter, said plurality of lines potentially having at least one skew relative to the boundaries of said image, requiring registration, containing inverted text matter, undesired dots, and vertical and horizontal lines, said process comprising

a step for providing a primary scan line run-length coded image of said document;

a step for determining said skew;

a step for reducing said skew;

a step for determining said registration;

a step for correcting said registration;

a step for detecting said inverted text matter;

a step for reversing said inverted text matter;

a step for detecting and mapping said undesired dots in an information removal map;

a step for deleting said undesired dots contained in said information removal map;

a step for detecting and mapping said vertical lines in said information removal map;

a step for eliminating said vertical lines contained in said information removal map;

a step for detecting and mapping said horizontal lines in said information removal map;

a step for eliminating said horizontal lines contained in said information removal map.

9. A process for enhancing an image of a scanned document as claimed in claim 8 wherein said step for deleting said undesired dots, said step for eliminating said vertical lines, and said step for eliminating said horizontal lines are executed as one step for removing said undesired dots, said vertical lines and said horizontal lines mapped in said information removal map.