Image analysis – Color image processing – Color correction
Reexamination Certificate
1999-08-18
2003-01-07
Mancuso, Joseph (Department: 2721)
Image analysis
Color image processing
Color correction
C382S257000, C382S258000, C382S259000
Reexamination Certificate
active
06504953
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention is in the field of electronic reproduction technology and is directed to a method for the automatic removal of errors in a digitally stored image.
In reproduction technology, print masters for printing pages are produced that contain all page elements such as texts, graphics and images to be printed. In the case of electronic production of the print masters, these elements are present in the form of digital data. For example, the data are generated for an image in that the image master is scanned point-by-point and line-by-line in a scanner, each picture element is resolved into color components, and the color values of these components are digitized. Dependent on the output process employed later, for example output on a color printer or printing in a conventional printing press, the data for the page elements are generated and stored in the color components red, green and blue (RGB) or in the inks of four-color printing cyan, magenta, yellow and black (CMYK).
During the course of further work, the digitized images together with the texts and graphics are electronically mounted at a computer work station under visual control, being mounted on a color monitor or automatically according to stored layout rules. The finished printing page is thereby converted into a data format suitable for the output and is stored. The printing page data for each of the inks (RGB or, respectively, CMYK) are referred to as color separation data. Printing plates for a conventional printing press are manufactured with the color separation data, or they are transmitted directly to a color printer or to a digital printing press and are printed out thereat.
It occurs that the image data of the scanned images exhibit faulty locations such as, for example, scratches in the original master or error locations that arise due to hairs, fibers, etc. that adhere to the image master during the scan process. Such errors must be corrected in the reproduction before the images can be printed. For that purpose, various electronic retouch methods are utilized in the prior art.
A known retouch method is copying retouch as disclosed by European Letters Patent 0 111 026. It is used in order to transfer information of an image area onto another image area picture element by picture element. Error locations in the image can thus be eliminated in that picture elements from a neighboring image area having similar color and structure are copied into the damaged image area. For implementing the retouch, the operator simultaneously moves a read mark and a write mark with a computer mouse at a computer workstation in which the image to be retouched is stored, said read mark and write mark being mixed on the picture screen. The read mark indicates a read area of the image, and the write mark indicates a write area in which the error location is located. The picture elements situated under the read mark in the stored image are thereby continuously transferred into the corresponding picture elements under the write mark.
Another method suitable for the elimination of error locations is known under the name of “bleeding retouch”. Two marks mixed into the picture screen are thereby likewise moved with the computer mouse by the operator. One mark is located at the one side of the error location, for example one side of a scratch, and the other mark is located at the opposite side of the error location. Two picture elements are read from the stored image that lie at the locations identified by the marks. Intermediate values are interpolated from the color separation values of the picture elements that are being read for all picture elements that lie on the connecting line between the, two marks and these intermediate values are written into the picture elements on the connecting line. In this way, the error location is covered with color values that are adapted to the colors in the neighborhood of the error location.
European Letters Patent 0 768 621 discloses a partially automated method for illuminating error locations. First, the operator manually identifies the error locations as a bit map in-a mask memory. The picture elements of the error locations identified in the bit map are then corrected with a specific window operator. The window operator places line segments over every picture element of the error location from different angles, whereby the outer ends of the line segments project into non-defective neighboring regions of the error location. The color values of the non-defective picture elements are compared along each line segment to color values that were acquired from an interpolation model. The line segment that exhibits the fewest interpolation errors is then selected. Finally, the defective pixel of the error location is interpolated along the selected line segment according to the interpolation model.
In European Letters Patent 0 614 154, the operator places a narrow limiting mask over an arrow location and the picture elements within the mask are colored white. The continuous color values in a processing area around the limiting mask are converted into a sequence of binary images, in that the color values are compared to thresholds that increase step-by-step. The contour of each binary image is diminished by an edge width that corresponds to the width of the limiting mask. The shape of the limiting mask is thereby eliminated. Subsequently, the modified shape of each binary image is again enlarged by the same edge width, and the sequence of binary images is converted back into an image having continuous color values. The error location is then eliminated from this image.
Methods known from the prior art for removing image errors have the disadvantage that an operator must first seek the position of the error locations in the image and identify them, this being extremely time-consuming. Moreover, error locations can thereby be overlooked by the operator. Subsequently, the operator must manually correct the error locations by interactive retouch in most methods, in that he moves marks mixed in on the picture screen over the error locations. This requires additional time expenditure.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to avoid the disadvantages of the known methods for removing image errors in digital image data and to specify a new, improved method that enables an automatic removal of error locations. Error locations can thus be removed from an image significantly faster then is possible with the known methods.
According to the method of the invention for automatic removal of image errors in digital image data, a contour mask is produced with a contour filter. The color mask is generated. An error mask is produced by a logical AND operation of the contour mask and the color mask. Contours in the error mask are pared down and vectorized. An automatic bleeding retouch of the digital image data is implemented over the vectorized contours of the error mask.
REFERENCES:
patent: 4868884 (1989-09-01), Miyazaki et al.
patent: 5134668 (1992-07-01), Appel
patent: 5619592 (1997-04-01), Bloomberg et al.
patent: 5659490 (1997-08-01), Imamura
patent: 5859929 (1999-01-01), Zhou et al.
patent: 5887082 (1999-03-01), Mitsunaga et al.
patent: 6167167 (2000-12-01), Matsugu et al.
patent: 6212287 (2001-04-01), Olivieri et al.
patent: 6404936 (2002-06-01), Katayama et al.
Heidelberger Druckmaschinen Aktiengesellschaft
Le Brian
Mancuso Joseph
Schiff & Hardin & Waite
LandOfFree
Method for the automatic removal of image errors 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 for the automatic removal of image errors, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for the automatic removal of image errors will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3012854