Image analysis – Image transformation or preprocessing – Combining image portions
Reexamination Certificate
2000-07-14
2003-07-01
Mehta, Bhavesh M. (Department: 2625)
Image analysis
Image transformation or preprocessing
Combining image portions
C345S629000, C348S586000
Reexamination Certificate
active
06587599
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of Invention
The present invention relates generally to image recognition and more particularly relates to methods and apparatuses for preserving background continuity in images when rendered objects in the images or other graphic objects are superimposed to the images.
2. Description of the Related Art
The Internet is a rapid growing communication network of interconnected computers and computer networks around the world. Together these millions of connected computers provide a new paradigm for information exchange via the Internet. Among the information being exchanged, images play an important role owing to its fidelity in preserving information across different platforms. Transmissions of image data, however, are always costly and slow because of the vast data involved and limited bandwidth in the Internet.
For images to be transmitted over a data network, a compressed version of the images must be obtained before the transmissions of the images take place. However, none of the current compression techniques are fully satisfactory because of either the poor resultant quality or low compression ratios. Efforts towards compression of such images are still going on. Nevertheless, there are a certain number of images that are graphics-based and can be transmitted efficiently without being compressed. Graphics-based images, unlike scenery images, can be analyzed to form a set of primitives from which the original graphics-based images can be exactly recovered. For example, an image of an article, instead of transmitting the image of the article, the text in the article is transmitted after the text is first recognized. Upon receiving the text, it is not difficult to reconstruct the image of the article.
It is well known that, besides text, graphic objects such as line, rectangle, circle and other geometric shapes can be presented by a small set of primitives with which the original shapes can be exactly recovered as well. Further there are applications that need to extract a set of primitives from images for archive. With derived primitives, images can be automatically searched based on the frequency of permutation of the primitives. One of most interested primitives from graphics-based images is to recognize texts. In the past ten years, several commercial optical character recognition (OCR) products, such as Caere's OmniPage and WordScan, Xerox's TextBridge, Expervision's TypeReader, and Adobe's Capture have been introduced and are capable of recognizing text in images of articles. However, there are no products that can fairly recover the images based on the recognized text. With today's printing technology, many texts are printed in a colorful background, to overlay the rendered text onto the colorful background while preserving continuity in the background remains a challenge problem. There is, therefore, a great need for a system that provides a generic solution to preserving background continuity when images are reconstructed with rendered objects.
SUMMARY OF THE INVENTION
The disclosed system has been made in consideration of the above described problems and has particular applications to image archives and reediting. Given an image of an article having graphic objects including texts and pictures, the disclosed system first distinguishes graphic objects from the background by segmenting the image so as to produce an object image. With respect to the object image, the disclosed system “repaints” those spaces previously occupied by the graphic objects, referred to as object spaces, with the background color as such the background continuity is preserved when the rendered objects are superimposed onto the repainted background.
According to one aspect of the invention, the disclosed system is a method for preserving the background continuity in an image when rendered objects are superimposed thereto. The method comprises:
acquiring the image comprising graphic objects;
producing an object image from the acquired image; wherein the object
image comprises a background and a foreground, the foreground
corresponding to the graphic objects in the acquired image; generating a patched image from the acquired image with respect to the
foreground in the object image;
receiving rendered objects corresponding to the foreground; and
reformatting the rendered object.
The method further comprises reformatting the rendered object, wherein the render objects may be generated from the acquired image, so that the composed output image containing the rendered objects resembles the acquired image.
According to another aspect of the invention, the disclosed system is an apparatus for preserving the background continuity in images. The apparatus comprises:
a working memory containing the image comprising graphic objects, each of the graphic objects occupying a respective object space;
a display screen;
an object engine generating an object image from the image, the object image containing a foreground and a background, the foreground corresponding to the graphic objects, respectively, and the object image displayed on the display screen; and
a patch generator generating a patched image from the object image, the patched image displayed in the display screen; and
an object rendering engine producing rendered objects.
The apparatus further comprises a composer superimposing rendered objects onto the object image, wherein the rendered objects are generated from the image with respect to the object image.
Accordingly, an important object of the present invention is to provide a generic solution to preserve background continuity when reconstructing images with rendered graphic objects.
REFERENCES:
patent: 5220312 (1993-06-01), Lumelsky et al.
patent: 5420971 (1995-05-01), Westerink et al.
patent: 5627959 (1997-05-01), Brown et al.
patent: 5745095 (1998-04-01), Parchem et al.
patent: 5754184 (1998-05-01), Ring et al.
patent: 5914748 (1999-06-01), Parulski et al.
patent: 5937104 (1999-08-01), Henderson et al.
Mehta Bhavesh M.
XMLCities, Inc.
Zheng Joe
LandOfFree
Method and apparatus for producing a composed output... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for producing a composed output..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for producing a composed output... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3048132