Handheld device document imaging

• US 8,817,339 B2
• Filed: 01/20/2013
• Issued: 08/26/2014
• Est. Priority Date: 08/22/2012
• Status: Active Grant

• Current Assignee: Top Image Systems Limited (Kofax Incorporated)
• Original Assignee: Top Image Systems Limited (Kofax Incorporated)
• Inventors: Hinski, Yuval
• Primary Examiner(s): Worku, Negussie
• Application Number:
  US13/745,806
• Publication Number:
  US 20140055826A1
• Time in Patent Office: 583 Days
• Field of Search: 358/1.9, 358/450, 358/453, 358/540, 358/474, 358/473, 382/284, 382/294, 382/309, 382/212, 382/213, 382/285, 382/282, 348/231.5, 348/159, 348/36, 348/239, 348/42, 348/218.1
• US Class Current: 358/474
• CPC Class Codes: H04N 1/3876 Recombination of partial images to recreate the original image
• CPC Class Codes: H04N 2201/0081 Image reader H04N2201/0091 - H04N2201/0094 take precedence
• CPC Class Codes: H04N 2201/0096 Portable devices

Abstract

A method of stitching frames of a video sequence to image a target document. The method comprises capturing a group of frames of a video sequence using an image sensor of a handheld device having a display, during the capturing, analyzing the video sequence to select iteratively a group of the frames, each member of the group depicts another of segments of a target document, during the capturing, sequentially presenting a plurality maneuvering indications, each the maneuvering indication is presented after a certain frame depicting a certain of the segments is captured and indicative of a maneuvering gesture required for bringing the image sensor to capture another frame depicting another segment of the segments, the another segment being complementary and adjacent to the certain segment, and stitching members of the group to create a mosaic image depicting the target document as a whole.

Claims

1. A method of stitching a plurality of frames of a video sequence to image a target document, comprising:

capturing a group of a plurality of frames of a video sequence using an image sensor of a handheld device having a display;

during said capturing, analyzing said video sequence to select iteratively a group of said plurality of frames, each member of said group depicts another of a plurality of segments of a target document;

during said capturing, sequentially presenting a plurality of visual maneuvering indications on said display, while said display presents said plurality of frames, each said visual maneuvering indication is presented after a frame depicting a segment of said plurality of segments of said target document is captured and said visual maneuvering indication is indicative of a maneuvering gesture required for bringing said image sensor to capture another frame depicting another segment of said plurality of segments, said another segment being complementary and adjacent to said previous segment; and

,

stitching members of said group to create a mosaic image depicting said target document as a whole.

2. The method of claim 1, wherein said members are selected sequentially, each said member is stitched to another member before a further another member is selected.

3. The method of claim 1, further comprising identifying at least one dimension of said target document, wherein said plurality maneuvering indications are set to instruct said user to maneuver said handheld device along a scanning pattern that is adapted to said at least one dimension.

4. The method of claim 1, wherein said analyzing comprises detecting a first frame depicting a corner of said document and using said first frame to generate at least one of said plurality of maneuvering indications.

5. The method of claim 1, wherein said presenting comprises presenting instructions to align said handheld device in relation to said target document.

6. The method of claim 1, further comprising estimating a current motion direction of said handheld device during said capturing;

wherein each member of said group is stitched to another member during said capturing according to said current motion direction.

7. The method of claim 1, wherein said plurality of segments comprises said mosaic image when arranged in a crisscross manner, one row of segments on top of the other.

8. The method of claim 7, wherein said analyzing comprises detecting a first frame depicting a corner of said document and using said first frame as a trigger to generate a maneuvering indication of said plurality of maneuvering indications which indicates to a user to capture a second frame depicting a segment from another row of segments.

9. The method of claim 7, wherein said plurality of visual maneuvering indications comprise a plurality of arrows which are sequentially presented on a display of said handheld device to indicate to said user to guide said handheld device along a scanning pattern.

10. The method of claim 1, further comprising automatically identifying a plurality of characters in said mosaic image.

11. The method of claim 1, wherein said analyzing comprises selecting said certain frame after removing a blur therefrom.

12. The method of claim 1, wherein said analyzing, said presenting, and said stitching are automatically performed.

13. The method of claim 1, wherein said stitching comprises normalizing said mosaic image.

14. The method of claim 1, wherein said stitching comprises binarizing said mosaic image.

15. A non-transitory computer readable storage medium comprising computer executable instructions adapted to perform the method of claim 1.

16. A handheld device of stitching a plurality of frames of a video sequence to image a document, comprising:

an image sensor which captures a group of a plurality of frames of a video sequence;

a display;

a processor;

an analysis module which uses said processor to analyze said video sequence to select iteratively a group of said plurality of frames, each member of said group depicts another of a plurality of segments of a document;

a maneuvering indication module which sequentially presents a plurality of visual maneuvering indications on said display, while said display presents said plurality of frames, during said capturing, each said visual maneuvering indication being presented on said display after a frame depicting a segment of said plurality of segments of said document is captured and indicative of a maneuvering gesture required for bringing said image sensor to capture another frame depicting another segment of said plurality of segments, said another segment being complementary and adjacent to said previous segment; and

a stitching module which stitches said plurality of segments to create a mosaic image depicting said document as a whole.

17. The handheld device of claim 16, wherein said handheld device is a cellular phone.

18. A non-transitory computer readable storage medium encoded with computer executable instructions for performing steps including stitching a plurality of frames of a video sequence to image a document, said steps comprising:

capturing a group of a plurality of frames of a video sequence using an image sensor of a handheld device having a display;

analyzing, during said capturing, said video sequence to select iteratively a group of said plurality of frames, each member of said group depicts another of a plurality of segments of a document;

presenting sequentially, during said capturing, a plurality of visual maneuvering indications, on said display, while said display presents said plurality of frames, each said visual maneuvering indications is presented after a certain frame depicting a certain of said plurality of segments is captured and indicative of a maneuvering gesture required for bringing said image sensor to capture another frame depicting another segment of said plurality of segments, said another segment being complementary and adjacent to said certain segment; and

,

stitching said plurality of segments to create a mosaic image depicting said document as a whole.