Image analysis – Image enhancement or restoration – Intensity – brightness – contrast – or shading correction
Reexamination Certificate
1999-06-30
2002-04-23
Boudreau, Leo (Department: 2621)
Image analysis
Image enhancement or restoration
Intensity, brightness, contrast, or shading correction
C382S167000, C382S199000, C382S205000, C382S218000, C382S266000, C382S284000, C358S001200, C358S001900
Reexamination Certificate
active
06377711
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of Invention
This invention generally relates to processing a color image.
2. Description of Related Art
Images are stored as discrete picture elements, or pixels, with each pixel corresponding to an identified image point and represented by a pixel value. An entire set of pixels representing an image is referred to as a bitmap (for binary pixel values) or a bytemap (for multi-bit pixel values) of the image. Color pixel values generally include information such as values for attributes such as hue, color and saturation. This information can be supplied directly to display devices, such as cathode ray tube (CRT) displays, that can directly represent each pixel's information.
However, polychromatic or color printing generally involves separately printing successive layers of colored inks. Each layer of ink is applied over the previous layer. The number of stages for printing an image depends upon the color model chosen. In the most common color model, subtractive color printing, separate cyan (c), magenta (m), yellow (y) and sometimes black (k) ink layers are successively applied.
Thus, in order to represent a polychromatic image, the image is first broken down into a set of monochromatic images, each of which corresponds to the contribution of a particular color (cyan, magenta, yellow or black) to the overall image. The quality of the image depends on the mechanical precision with which the printer registers the successive ink layers.
SUMMARY OF THE INVENTION
A raster is a two-dimensional pattern, divided into a discrete number of positions that cover an image area. Trapping is a technique of intentional overlapping adjacent colored objects on a page to cover up printer registration errors. Printer misregistration can occur in the horizontal, vertical and diagonal directions. Any trapping solutions must provide overlap in edges along each of these directions.
In conventional trapping methods, diagonal, vertical and/or horizontal edges between pixels are located and then the pixel values are adjusted to reduce or eliminate the effects of misregistration of the various colored layers with respect to each other.
The following U.S. patents, each of which is incorporated herein by reference in its entirety, disclose methods of detecting locations to trap based on raster data: U.S. Pat. Nos. 5,615,314; 5,513,300; 5,440,652; 5,386,483; 5,386,223; 5,295,236; 5,241,396; 5,131,058; 4,931,861; 4,725,966; 4,700,399; and 4,583,116. For example, U.S. Pat. No. 4,700,399 to Yoshida discloses a color image processing apparatus including a detector for detecting an edge of an image. The Yoshida edge detector detects edges in yellow, magenta and cyan color signals using a Laplacian algorithm. An edge quantity is calculated as the sum of differences between a density level for a pixel and the densities of the eight adjacent pixels.
Known raster-based trapping solutions only look for edges between each pixel and its four vertically and horizontally connected neighbors (e.g., horizontal and vertical edges).
FIGS. 1 and 2
illustrate the effect of trapping only horizontal and vertical edges. In
FIG. 1
, a conventional edge detecting system scans for edges between a central or target pixel, or a pixel of interest,
22
and adjacent or neighboring pixels
12
,
21
,
23
and
32
. An edge between the target pixel
22
and one of the adjacent pixels
11
,
13
,
31
and/or
33
would not be detected. This can lead to corners being knocked out of the trap overlap region, allowing “white” space to be shown under diagonal misregistration. This is illustrated in
FIG. 2
, which shows a processed set of pixels including an object
122
bounded by an edge
141
. As can be seen, the processing has left the corners
143
-
146
“unprocessed”.
An improved system achieves horizontal, vertical and diagonal trapping by looking for edges between pixels, such as the target pixel
22
, and the eight connected neighboring pixels, such as the pixels
11
,
12
,
13
,
21
,
23
,
31
,
32
and
33
, that are adjacent to the target pixel
22
, as shown in
FIGS. 3 and 4
. This conventional system removes the possibility for missing corners, as can occur in the example shown in
FIGS. 1 and 2
. However, the improved system requires additional time to conduct the edge detecting routines. This additional time is considerable when detecting edges for each pixel in the image.
Thus, the known approaches to edge detection, including those represented in
FIGS. 1-4
, have been found to be excessive in most situations.
This invention provides systems and methods for trapping raster images using only diagonal edge detection.
This invention separately provides methods and systems that look for edges between pixels by selecting a target pixel and looking at two neighboring pixels adjacent to the selected pixel.
This invention separately provides methods and systems which perform edge detecting routines much faster than the conventional edge detecting approaches.
These and other features or advantages of this invention are described in or are readily apparent from the following detailed description of the various exemplary embodiments.
REFERENCES:
patent: 5333211 (1994-07-01), Kanda et al.
patent: 5542052 (1996-07-01), Deutsch et al.
patent: 5581667 (1996-12-01), Bloomberg
patent: 5613046 (1997-03-01), Dermer
patent: 5956157 (1999-09-01), Tai
patent: 6141462 (2000-10-01), Yoshino et al.
patent: 6195467 (2001-02-01), Asimopoulos et al.
Boudreau Leo
Mariam Daniel G.
Oliff & Berridg,e PLC
LandOfFree
Methods and systems for detecting the edges of objects in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and systems for detecting the edges of objects in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and systems for detecting the edges of objects in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2856799