Facsimile and static presentation processing – Facsimile – Recording apparatus
Reexamination Certificate
1998-12-14
2001-11-13
Rogers, Scott (Department: 2624)
Facsimile and static presentation processing
Facsimile
Recording apparatus
C358S463000
Reexamination Certificate
active
06317223
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an image processing system which removes vertical artifacts such as lines on images produced by scanning an original document, without loss of image information.
BACKGROUND OF THE INVENTION
Unexpected vertical black patterns or artifacts such as lines, dots, dashes, etc., have been a common image artifact in binary document images produced by production paper scanners. The vertical black patterns are not part of the original document image, however, the patterns are the result of pieces of dust or dirt blocking light between an illumination light source and the paper documents being scanned. For example, pieces of dust or dirt which come to rest on transparent image guides in an illumination path of the scanner will result in shaded lines upon the document moving through the scanner. When the document is electronically captured by a CCD device, the shaded lines will appear in the digital document image. The shaded vertical lines within a grey scale image will be indistinguishably treated as part of image content. After an adaptive thresholding is applied to the grey scale image, the shaded lines will show up as black lines in the binary image.
One current solution to this problem is to routinely remove the dust or dirt from the illumination path by cleaning the transparent image guides. However, this solution requires interruption of the scanner operation which is undesirable in a production scanner.
SUMMARY OF THE INVENTION
An object of the present invention is to provide for a digital image processing method which identifies vertically disposed patterns or artifacts caused by dirty image guides in scanners and then subtracts the vertically disposed patterns before creating the final image of the scanned document. The vertically disposed patterns can be in the form of lines, dashes, dots, etc.
The present invention provides for a digital image processing method which includes the steps of (1) applying an edge-based adaptive thresholding method twice with different contrast parameter settings to convert an image such as a gray scale image into two binary images; (2) generating a map of the difference from the two binary images; (3) projecting the vertical profile of the map; (4) detecting the local peaks, (i.e. vertical lines, dashes, dots, etc.) from the vertical projection profile; (5) locating the regions of the vertical lines, dashes, dots, etc. in the map based on the detected local peaks; and (6) reversing pixels from the binary images based on the vertical lines, dashes, dots, etc. of the map.
The method of the present invention includes: capturing a document image by a CCD array in raster scan fashion; applying an edge-based adaptive thresholding technique [1] with a normal contrast setting to extract every detail of image information and storing the resulting binary image, named as B
1
; applying the same edge-based adaptive thresholding technique with a low contrast setting which only extracts high contrast details of the image, named as B
2
; labelling the difference of B
1
and B
2
, and storing the map of the difference, named as D; projecting the vertical profile of the map (D); locating the vertically disposed patterns by detecting the local peaks in the vertical projection profile, resulting in a map of the vertically disposed patterns, named as V; reading the binary image (B
1
) and the map of the vertically disposed patterns (V); and reversing pixels in the B
1
image at the corresponding black pixel locations in the V map.
The present invention provides for an image processing method which comprises the steps of: converting an image into first and second binary images; comparing the first and second binary images and generating a map indicative of a difference between the first and second binary images; projecting a vertical profile of the map and detecting peaks in the vertical profile; locating regions of vertically disposed patterns in the map based on the detected peaks; and reversing pixels from the first binary image based on the vertically disposed patterns in the map.
The present invention further relates to an image processing method which comprises the steps of: capturing an image; extracting image information from the image and storing a resulting image as a first binary image; extracting high contrast details from the image and storing a resulting image as a second binary image; comparing the first and second binary images and generating a bitmap indicative of a difference between the first and second binary images; projecting a vertical profile of the bitmap; locating regions of vertically disposed patterns in the bitmap by detecting local peaks in the vertical profile so as to provide for a map of vertically disposed patterns; and reversing pixels from the first binary image based on the vertically disposed patterns of the map.
The present invention further relates to an image processing method which comprises the steps of: converting an image into first and second binary images; generating a map indicative of a difference between the first and second binary images; and reversing pixels from the first binary image based on vertically disposed patterns in said map.
The present invention further relates to an image capture assembly that comprises an image capture section which captures an image; a conversion section which converts the captured image into digital image information indicative of the captured image; and a processing section which receives the digital information and converts the captured image into first and second binary images. The processing section compares the first and second binary images to generate a map indicative of a difference between the first and second binary images and reverses pixels from the first binary image based on vertically disposed patterns in the map.
This invention takes advantage of a general trend that any real-world document would not contain significant “page length” vertical lines of low contrast. Typically, only variable date (handprint, typewriter etc.) would be of low contrast, while lines and graphics would be part of a form (high contrast).
REFERENCES:
patent: 4649436 (1987-03-01), Nakagawa et al.
patent: 5214470 (1993-05-01), Denber
patent: 5223951 (1993-06-01), Umakoshi
patent: 5436979 (1995-07-01), Gray et al.
patent: 5583659 (1996-12-01), Lee et al.
patent: 5974194 (1999-10-01), Hirani et al.
patent: 6035072 (2000-03-01), Read
patent: 6125213 (2000-09-01), Morimoto
patent: 1014677 A2 (2000-06-01), None
patent: 1014691 A2 (2000-06-01), None
Lee Yongchun
Rudak Peter
Stefanko James M.
Eastman Kodak Company
Rogers Scott
LandOfFree
Image processing system for reducing vertically disposed... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing system for reducing vertically disposed..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing system for reducing vertically disposed... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2578117