Image processing device, image processing method, scanner and storage medium

Patent metadata

Field	Value
Publication number	US-9239954-B2
Application number	US-201313906854-A
Country	US
Kind code	B2
Filing date	May 31, 2013
Priority date	May 31, 2012
Publication date	Jan 19, 2016
Grant date	Jan 19, 2016

Abstract

The disclosure provides an image processing device, image processing method, scanner and storage medium. The image processing device is used for tracing a boundary of an object image in an image, the boundary being continuous and the rate of change in slope between adjacent points on the boundary being slow. The image processing device includes: a boundary estimation unit, adapted to estimate the location of the boundary of the object image; an interfering gradient processing unit, adapted to process an interfering gradient near the estimated boundary, so as to reduce the interfering gradient or remove the interfering gradient from the image; and a boundary tracing unit, adapted to trace the boundary in the image having the interfering gradient processed. By using the technique of the disclosure, the accuracy of tracing a boundary of an image is improved significantly.

First claim

The invention claimed is:

1. An image processing device for tracing a boundary of an object image in an image, the boundary being continuous and the rate of change in slope between adjacent points on the boundary being slow, the image processing device comprising:
a boundary estimation unit, adapted to estimate the location of the boundary of the object image;
an interfering gradient processing unit, adapted to process an interfering gradient near the estimated boundary, so as to reduce the interfering gradient or remove the interfering gradient from the image; and
a boundary tracing unit, adapted to trace the boundary in the image having the interfering gradient processed;
wherein the boundary estimation unit comprises:
a segmentation unit, adapted to segment a portion of the image between two corners on the boundary into a predetermined number of segments;
a segment boundary estimation unit, adapted to estimate the location of the boundary respectively in each of the segments;
wherein the segment boundary estimation unit comprises a boundary approximation unit for approximating the boundary in each of the segments using a straight line segment,
wherein the boundary approximation unit is adapted to, for at least one of outermost segments where the corners are in respectively, set the corner in the outermost segment as a start point of a first straight line segment used to approximate the boundary in the outermost segment, and select an end point for the first straight line segment on a first segmentation line resulting in the outermost segment, so that in a gradient histogram of a narrow band covering the first straight line segment, pixels exhibit the greatest degree of concentration with respect to the greatest gradient; and
the boundary approximation unit is further adapted to set the obtained end point of the boundary-approximating straight line segment as a start point of a second straight line segment used to approximate the boundary in a next, adjacent segment, and select an end point for the second straight line segment on a second segmentation line which results in the next segment and is not common to the next segment and the segment for which the boundary-approximating straight line segment has been obtained, so that in a gradient histogram of a narrow band covering the second straight line segment, pixels exhibit the greatest degree of concentration with respect to the greatest gradient.

2. The image processing device according to claim 1 , wherein the segment boundary estimation unit further comprises an evaluation unit adapted to evaluate usability of an approximating straight line segment obtained by the boundary approximation unit.

3. The image processing device according to claim 2 , wherein the evaluation unit comprises: a concentration degree evaluation unit, adapted to determine whether a degree at which the pixels concentrate on the greatest gradient is larger than or equal to a predetermined degree, and if it is larger than or equal to the predetermined degree, the concentration degree evaluation unit determines that the approximating straight line segment of the boundary is available; otherwise, the concentration degree evaluation unit determines the approximating straight line segment of the boundary is unavailable.

4. The image processing device according to claim 2 , wherein the evaluation unit comprises: an inclination angle evaluation unit, adapted to compare an inclination angle of a newly-obtained approximating straight line segment with an inclination angle of an approximating straight line segment in a previous, adjacent segment, and if the difference between the inclination angles is smaller than or equal to a predetermined threshold, the inclination angle evaluation unit determines the approximating straight line segment is available; otherwise, the inclination angle evaluation unit determines the approximating straight line segment is not available.

5. The image processing device according to claim 1 , wherein the interfering gradient processing unit is adapted to process an interfering gradient near the approximating straight line segment for each of the segments.

6. The image processing device according to claim 5 , wherein, the interfering gradient processing unit processes gradient in a predetermined region except for a narrow band covering the approximating straight line segment.

7. The image processing device according to claim 2 , wherein, when an approximating straight line segment in the outermost segment is unavailable, the interfering gradient processing unit is adapted to process gradients except for the narrow band covering a predetermined width above and under a horizontal line where corners are in; and the boundary tracing unit is adapted to trace a boundary of an object image of the image subjected to the process.

8. The image processing device according to claim 2 , wherein, when approximating is only started from a corner at one end, and an approximating straight line segment of a certain segment is unavailable, the interfering gradient processing unit is adapted to connect an end point of an approximating straight line segment of the last segment for which the approximating is available to a corner at the other end, and to process the gradient of the connecting line at the side where the boundary does not exist.

9. The image processing device according to claim 2 , wherein, when approximating is started simultaneously from corners at both ends, and an approximation of a segment at a certain side or at both sides is unavailable, the interfering gradient processing unit is adapted to connect an end point of an approximating straight line segment of the last segment for which the approximating is available to an end point of an approximating straight line segment of the last segment for which the approximating is available at the other side, and to process the gradient of the connecting line at the side where the boundary does not exist.

10. The image processing device according to claim 1 , wherein the object image is a document image.

11. A scanner, comprising the image processing device according to claim 1 .

12. An image processing method of tracing a boundary of an object image in an image, the boundary being continuous and the rate of change in slope between adjacent points on the boundary being slow, the image processing method comprises:
a boundary estimation step, of estimating the location of the boundary of the object image;
an interfering gradient processing step, of processing an interfering gradient near the estimated boundary, so as to reduce the interfering gradient or remove the interfering gradient from the image; and
a boundary tracing step, of tracing the boundary in the image having the interfering gradient processed; wherein, the boundary estimation step comprises:
a segmentation step of segmenting a portion of the image between two corners on the boundary into a predetermined number of segments;
a segment boundary estimation step of estimating the location of the boundary respectively in each of the segments;
wherein the segment boundary estimation step comprises a boundary approximating step of approximating the boundary in each of the segments using a straight line segment, the boundary approximating step comprising:
for at least one of outermost segments where the corners are in respectively, setting the corner in the outermost segment as a start point of a first straight line segment used to approximate the boundary in the outermost segment, and selecting an end point for the first straight line segment on a first segmentation line resulting in the outermost segment, so that in a gradient histogram

Assignees

• Fujitsu Ltd

Inventors

• Pan Pan
• He Yuan
• Sun Jun
• Naoi Satoshi

Classifications

• G06V30/18076

  by analysing connectivity, e.g. edge linking, connected component analysis or slices · CPC title

• G06V30/146 Primary

  Aligning or centring of the image pick-up or image-field · CPC title

• G06V30/414

  Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text · CPC title

• G06V30/10

  Character recognition · CPC title

• G06T2207/30176

  Document · CPC title