Computer graphics processing and selective visual display system – Display peripheral interface input device – Light pen for fluid matrix display panel
Reexamination Certificate
1998-07-20
2001-05-15
Chang, Kent (Department: 2673)
Computer graphics processing and selective visual display system
Display peripheral interface input device
Light pen for fluid matrix display panel
C345S182000, C358S451000
Reexamination Certificate
active
06232953
ABSTRACT:
FIELD OF INVENTION
This invention relates to optical scanners in general and more specifically to a method and apparatus for improving the image quality of scanned halftone images.
BACKGROUND
Most photographic images that are reproduced on paper are reproduced using a halftone printing process. For example, a black and white image is reproduced by the halftone printing process by photographing the image through a fine screen or grid. The purpose of the screen is to divide the continuous photographic image into a plurality of tiny dots, the darkness and density of which effectively reproduce the various shades of gray contained in the original photograph. The screen is generally selected so that the image dots are so small that they cannot be resolved by the eye when the halftone image is viewed at a normal distance. Consequently, the halftone image appears to be continuous, rather than a large number of individual dots.
The screen that is used to create the tiny halftone image dots is generally rectilinear so that the resulting image dots occur at regular intervals and are arranged along rows that are evenly spaced from one another. The screen is also usually placed over the original photograph at an angle to reduce the chances of creating interference patterns, such as moire patterns, with features in the image. The spacing between the rows of dots is referred to herein as the “halftone period, ” and the angle at which the rows of dots are inclined with respect to the “vertical” side of the image is referred to as the “halftone angle.”
Halftone images are commonly used and may appear in a wide variety of printed media, such as newspapers, magazines, books, and comic books, just to name a few. For example, a black and white halftone image appearing in a newspaper may have about fifty (50) rows of dots per inch (i.e., a halftone period of 1/50), inclined at an angle of about 45° (the halftone angle) with respect to the “vertical” side of the image. Of course, higher quality black and white images, such as those found in books, are possible by decreasing the halftone period and dot size. High quality halftone images may have up to 200 rows of dots per inch (i.e., a halftone period of 1/200).
Color images may also be reproduced with the halftone process, with common examples being color comics appearing in the Sunday papers and, of course, the higher quality color images typically found in books. In the color halftone process, the color image comprises three sets of dots of different colors, typically the subtractive primary pigments of yellow, cyan, and magenta. Black may also be added to the image to secure more perfectly black features. As is the case for black and white halftone images, the dots of each respective color in a color halftone image are arranged in rows of dots. However, the rows of dots for each respective color are typically inclined at different angles with respect to the vertical. For example, the yellow dots may be inclined at an angle of 22.5°, the cyan dots at 45°, and the magenta dots at 67.5°. The black dots may be inclined at an angle of about 45° or about −22.50°.
Optical scanners are devices that produce machine (e.g., computer) readable data which are representative of a scanned object, such as a page of printed text or a halftone image. Briefly, a typical optical scanner uses a line-focus system to focus light reflected by the image onto the surface of a detector, such as a linear photosensor array that comprises a plurality of tiny light sensitive elements arranged along a line. Each element in the array corresponds to a small area location on the image that is commonly referred to as a “picture element” or “pixel,” and produces a data signal that is representative of the intensity of light from the pixel. Since the detector typically comprises a one dimensional array of photoelements arranged along a line, the electronic signals produced by the elements in the array together correspond to a small slice or line on the image that is commonly referred to as a “scan line.” Accordingly, the entire image may be scanned by moving the optical system with respect to the image so that the elements of the detector are successively exposed to adjacent scan lines on the image. As the image is scanned, then, data signals from the photoelements of the detector are received and processed by an appropriate data processing system which may subsequently store the data on a suitable medium or generate a display signal therefrom, allowing the image of the object to be reproduced on a display device such as a CRT or a printer.
While it is possible to scan a halftone image with an optical scanner of the type described above, the halftone image dots may create a number of image “artifacts” that detract significantly from the appearance and usability of the scanned image. For example, at certain scanning resolutions, the image pixels and halftone dots may interfere, creating moiré patterns which, if severe, may make the scanned image unusable. If the scanning resolution is adjusted to eliminate such moire patterns, the halftone image dots themselves may appear in the scanned image, again reducing the quality of the image.
While several methods have been developed in attempts to remove such image artifacts, they have tended to be somewhat less than totally effective, or have required computationally intensive algorithms. For example, methods have been developed that utilize two-dimensional frequency analysis methods, such as fast Fourier transforms (FFT), to determine the halftone period and angle, and then use simple averaging algorithms to “filter” the image data. Unfortunately, however, such FFT methods are computationally intensive and tend to be relatively slow or require powerful computers. Also, the simple averaging algorithms that are used to “filter” the image data tend to be inaccurate, which may create other kinds of image distortions.
Consequently, a need exists for a method and apparatus for removing artifacts from a scanned halftone image that does not rely on computationally-intensive algorithms, such as fast Fourier transforms, which can require significant processing time, thus slow down the artifact removal process. Such an improved system would desirably also achieve increased image reproduction accuracy, but without compromising the ability to remove substantially all of the artifacts from the halftone image.
SUMMARY OF THE INVENTION
The invention may comprise a method for removing artifacts from a bitmap of image data produced by scanning a halftone image. The halftone image has a plurality of image dots arranged along a plurality of halftone lines. The halftone lines are oriented at a halftone angle and spaced at a halftone period. The bitmap of image data includes a plurality of pixels, each pixel having an intensity value. The method of removing artifacts may comprise the steps of: (a) calculating a weighted average intensity value for a target pixel based on a weight factor for the target pixel and on weight. factors for neighboring pixels in the bitmap of image data; (b) repeating step (a) for each pixel in the bitmap of image data; and (c) replacing the intensity value of each pixel in the bitmap of image data with the weighted average intensity value for each such pixel.
The invention may also comprise a method of determining the halftone angle and the halftone period of the halftone image. This method may comprise the steps of: selecting a plurality of pixels from the bitmap of image data corresponding to m rows of n columns of pixels; adding together the intensity values of the m pixels in each of the n columns of pixels to create a SumProfile for each of the n columns of pixels and for each of a predetermined number of ColumnOffset values; adding together the SumProfiles for each of the ones of the ColumnOffset values to create a TotalChange corresponding to each of the predetermined number of ColumnOffset values; finding the ColumnOffset value that corresponds to a greatest TotalChange, wherein the ColumnOffset value for the greatest
Graham James J.
Veazey Judson
Yancey Michael J.
Chang Kent
Hewlett--Packard Company
LandOfFree
Method and apparatus for removing artifacts from scanned... 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 removing artifacts from scanned..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for removing artifacts from scanned... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2510511