Automatic image segmentation in the presence of severe...

Image analysis – Image enhancement or restoration – Edge or contour enhancement

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C382S286000

Reexamination Certificate

active

06704456

ABSTRACT:

This invention relates generally to detection of a particular images of several documents placed on the platen of an image input device in the presence of severe background bleeding, and more particularly to the automatic detection and segmentation of the series of images placed on the platen using the technologies of joining orphaned images, linear regression algorithms and collinearity detectors.
Cross reference is made to the following related applications are hereby incorporated by reference for their teachings:
“METHOD AND APPARATUS FOR IDENTIFYING A PLURALITY OF SUB-IMAGES IN AN INPUT IMAGE,” Dennis L. Venable et al., application Ser. No. 08/786,538, filed Jan. 21
st
, 1997;
“STRUCTURED IMAGE (SI) EDITOR AND METHOD FOR is EDITING STRUCTURED IMAGES,” Michael R. Campanelli et al., application Ser. No. 08/338,856, filed Nov. 14, 1994; and
“LOCATING THE POSITION AND ORIENTATION OF MULTIPLE OBJECTS WITH A SMART PLATEN,” Dennis L. Venable., application Ser. No. 08/785,109, filed Jan. 21
st
, 1997.
“DETECTING OVERLAPPING IMAGES IN AN AUTOMATIC IMAGE SEGMENTATION DEVICE WITH THE PRESENCE OF SEVERE BLEEDING,” Dennis L. Venable, application Ser. No. 09/412212, filed September, 1999.
BACKGROUND AND SUMMARY OF THE INVENTION
To improve the ease of use of scanning systems, in particular digital copiers and scanners, it is desirable to automatically detect and segment scanned objects (e.g., a hard copy document) on the platen of the input scanner. To facilitate this automated detection and segmentation, it is necessary to identify the position, shape and rotation angle of the object. Such a capability can enhance productivity by decreasing the time required for scanning documents, enable multiple images to be efficiently scanned and reduce the requirement for accurate presentation placed upon document feeding mechanisms.
The present invention is directed to a system intended to automatically determine the boundaries of multiple objects within a scanned image of a series of documents. The present invention combines a number of graphics and image processing techniques into an automated, system that provides productivity enhancement for digital copiers and scanning systems.
The present invention accomplishes these objectives by: 1) locating each of the objects within an input image; 2) modeling the shape of the identified object (e.g., rectangle); and 3) positioning the object in a digital image at a predefined locations.
Heretofore, a number of patents and publications have disclosed segmentation and detection, the relevant portions of which may be briefly summarized as follows:
U.S. Pat. No. 5,485,568 to Venable et al., issued Jan. 16, 1996, and hereby incorporated by reference, discloses a method and apparatus for representing a complex color raster image as a collection of objects in a structured image format—a hierarchical, device-independent format. A structured image document, generated using the techniques described by Venable, is a representation of data that may be rendered into a raster image. The data includes simple raster images as well as a hierarchical collection of sub-objects and raster processing operations. The possible data types for objects in the structured image include a raster image, text, graphics, image processing description, and files containing multiple image representations
U.S. Pat. No. 5,528,387 to Kelly et al., issued Jun. 18, 1996, teaches electronic image registration in a scanner. In particular, the edge data of a document is detected and skew angle calculated. The image is then rotated based upon the skew angle and non-image areas are filled using an image generation feature.
U.S. Pat. No. 4,922,350 to Rombola et al., issued May 1, 1990, discloses a two-pass scanning apparatus for detecting the size and position of an original document on a scanner platen. Image signals captured on a first scan are employed to determine boundaries and a best-fit magnification so that the image may be fit to a recording sheet using image signals generated on a subsequent scanning pass.
U.S. Pat. No. 5,253,765 to Moorehead et al, issued Oct. 19, 1993, teaches a system for sorting randomly sized objects (e.g., mushrooms). Invariant moments are employed, utilizing the complete pixel information for all pixels within the border of a captured image, to extract information about the mushroom size and orientation.
U.S. Pat. No. 5,220,398 to Horn et al. teaches an analog VLSI microchip that uses moments to determine the position and orientation of an object in a scene.
In “Invariant Fitting of Planar Objects by Primitives,” published in 1996
IEEE Proceedings of ICPR
'96, pp. 508-512 Voss et al. teach a method of pattern recognition using primitives such as triangles, rectangles, circles ellipses, super-quadratics, etc. The authors further describe a technique for describing the primitives using moments in a normalized manner; resulting in a decrease in the numerical effort.
In “MANAGING AND REPRESENTING IMAGE WORKFLOW IN PREPRESS APPLICATIONS”,
Technical Association of the Graphic Arts
(
TAGA
) Vol. 1, 1995 Proceedings pp. 373-385, hereby incorporated by reference for its teachings, Venable et al. teach the use of structured images to manage workflow.
In accordance with the present invention, there is provided a method for processing a digital image using an imaging input device so as to reduce bleeding of the edges of the detected images comprising receiving a digitized image including at least one of a plurality of representations of objects; and processing said digitized image into at least one of a plurality of images of said digitized image wherein said processing step includes defining a plurality of bins having a plurality of edge points making up each of said plurality of images of said digitized image.
Further, in accordance with the present invention there is provided a method for processing a digital image using an imaging input device reducing bleeding of the edges of the detected images by determining edge contours to generate bounding boxes comprising: identifying bins having edges that approximate 90 degrees, and modeling a rectangular image fitting for bins that approximate the 90 degrees edges.
Further, in accordance with the present invention there is provided a method for processing a digital image comprising the steps of: receiving a digitized image of a plurality of representations of objects; and processing said digitized image into a plurality of images representing said digitized image wherein said processing step includes generating a plurality of bins wherein said bins have a plurality of edge points making up the boundaries of each of said plurality of images of said digitized image.
Finally, in accordance with the present invention there is provided a method for processing a digital image using an imaging input device reducing bleeding of the edges of the detected images by determining edge contours to generate bounding boxes comprising: identifying bins on said edge contours having edges that approximate 90 degrees, and modeling a rectangular image for each the identified bins.
The techniques described herein enable a user to expediently scan a document or other object, automatically recognizing the shape of the document within the digitized image, and composing a digital document incorporating the object.
The techniques described above are advantageous because they improve the efficiency of a scanning process, in the presence of severe bleeding.


REFERENCES:
patent: 4922543 (1990-05-01), Hugo et al.
patent: 5220398 (1993-06-01), Horn et al.
patent: 5253765 (1993-10-01), Moorehead et al.
patent: 5313570 (1994-05-01), Dermer et al.
patent: 5418574 (1995-05-01), Miyabata et al.
patent: 5485568 (1996-01-01), Venable et al.
patent: 5528387 (1996-06-01), Kelly et al.
patent: 5974169 (1999-10-01), Bachelder
patent: 6088473 (2000-07-01), Xu et al.
patent: 6240208 (2001-05-01), Garakani et al.
patent: 6360021 (2002-03-01), McCarthy et al.
1996 IEEE Proceedings of ICPR 96' PP508-512 Voss et al, Invariant Fitting of Planar Objective by Primi

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Automatic image segmentation in the presence of severe... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Automatic image segmentation in the presence of severe..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic image segmentation in the presence of severe... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-3194331

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.