Image analysis – Image transformation or preprocessing – Changing the image coordinates
Reexamination Certificate
1998-03-24
2004-01-13
Mehta, Bhavesh M. (Department: 2625)
Image analysis
Image transformation or preprocessing
Changing the image coordinates
C358S001200
Reexamination Certificate
active
06678426
ABSTRACT:
FIELD OF THE INVENTION
This invention relates in general to imaging systems and print resolution enhancement and, more particularly, to mapping of lower resolution digital data to a higher resolution for subsequent printing on a lower resolution print engine.
BACKGROUND OF THE INVENTION
Electrophotographic processes for producing a permanent image on media are well known and commonly used. In general, a common process includes: (1) charging a photoreceptor such as a roller or continuous belt bearing a photoconductive material; (2) exposing the charged area to a light image to to produce an electrostatic charge on the area in the shape of the image; (3) presenting developer particles (toner) to the photoreceptor surface bearing the image so that the particles are transferred to the surface in the shape of the image; (4) transferring the particles in the shape of the image from the photoreceptor to the media; (5) fusing or fixing the particles in the shape of the image to the media; and (6) cleaning or restoring the photoreceptor for the next printing cycle. Many image forming apparatus, such as laser printers, copy machines, and facsimile machines, utilize this well known electrophotographic printing process.
In laser printers, an image is typically rasterized to form a bit pattern which is stored as a binary image bitmap for subsequent rendering to a final output image. The image bitmap is also referred to as a picture element (“pixel”) raster image. In the rasterizing process (i.e., forming the binary bitmap), graphic elements, such as continuous lines (line art) and text character outlines are converted to pixel patterns that approximate the source image shape. Continuous tone data, such as photographic data (both color and gray value images) are also converted to pixel patterns that approximate the source continuous tone image data. However, to effectively portray the original source image for continuous tone data, each pixel of the source image must be represented by multiple bits which define either a color or a gray level and which are subsequently converted, typically, to a binary image bitmap. Hereafter, it is to be understood that when the term “gray” is used, it applies to both color and black/white images and, when applied to a color image, relates to the intensity of the color.
Conventionally, in order to represent gray level images on a bi-level (black and white) printer, the pixel data, if not already gray level, is converted into a gray level, multi-bit configuration. For example, when a multi-bit configuration of 8 bits per pixel is employed, 256 gray levels can be represented by the digital pixel values. The individual gray level pixels are converted to binary level pixels (i.e., bi-level data for subsequent rendering) through the use of a dithering process. Spatial dithering (or digital halftoning) is the converting of the multi-bit pixel values (of a source image) to fixed-size, binary, multi-pixel groupings that approximate the average gray value of the corresponding source data. This dithering process provides a halftone texture to selected areas of the image so as to provide gray value variations therein. Thus, for example, with binary pixels, a 6×6 multi-pixel grouping can, in theory, simulate 36 levels of gray, and an 8×8 grouping can simulate 64 levels.
The dithering process (i.e., halftoning) employs a comparison of the individual pixel values (specified by a source image intensity array) against a threshold matrix (dither matrix or device best threshold array) to control the conversion of the gray level values to appropriate patterns of bi-level data. For purposes of this discussion, a gray level value of
255
in a source image is considered to be “white”, and a gray level value of 0 is “black”. The threshold matrix comprises a plurality of row-arranged gray level values which control the conversion of the gray level pixel values to bi-level pixel values which are stored in a resultant page buffer array (raster) bitmap. During the dithering process, the threshold matrix is tiled across the image pixels to enable each gray level image pixel to be compared against the correspondingly, logically-positioned gray level value of the threshold matrix. In essence, each entry in the threshold matrix is a threshold gray level value which, if exceeded by the source image gray level pixel value, causes that gray level image pixel to be converted to a “white” pixel (or, in this example, a binary logical “zero” for laser modulation purposes in the electrophotoconductive process). If, by contrast, the source image gray level pixel value is less than or equal to the corresponding threshold matrix gray level value, it is converted to a “black” pixel (or a binary logical “one” for laser modulation purposes).
Thus far, the discussion has focused on the differences between rasterizing text (or line art) and halftone images. However, in either case, once a raster page buffer array bitmap is generated from a source image, whether the image is text, line art, or halftone, the desired output image is created (rendered) by causing a laser to be modulated in accordance with the bit pattern stored in the image page buffer array bitmap. The modulated laser beam is scanned across a charged surface of a photosensitive drum in a succession of raster scan lines. Each scan line is divided into the pixel areas dictated by the resolution of the bitmap and the pitch of the laser scan. The modulated laser beam causes some pixel areas to be exposed to a light pulse and some not, thus causing a pattern of overlapping dots on each scan line. Where a pixel area (dot) is illuminated, the photosensitive drum is discharged, so that when it is subsequently toned, the toner adheres to the discharged areas and is repelled by the still charged areas. The toner that is adhered to the discharged areas is then transferred to paper and fixed in a known manner.
In general, the fidelity of the output image relative to the source data is directly related to the resolution of pixels (dots) in the output image. Arbitrary analog images cannot be exactly reproduced by a bitmap raster unless an infinite resolution is used. For example, as a result of the images's pixel configuration, image edges that are either not parallel to the raster scan direction or not perpendicular to it appear stepped. This is especially noted in text and line art.
Various techniques have been developed to improve the quality of the output image of a raster bitmap. These enhancement techniques include: edge smoothing, fine line broadening, antialiasing (to smooth jagged edges), and increasing the resolution of the laser printer. These enhancing techniques typically modify the signals to the laser to produce smaller dots that are usually offset from the pixel center, or in other words, to produce gray scale dots.
Although the prior art has attempted in a variety of ways to overcome the stepped appearance of pixel image edges for text and line art, an example of one of the more widely used techniques is described in U.S. Pat. No. 4,847,641 to Tung, the disclosure of which is incorporated in full herein by reference. Tung discloses a character generator that produces a bitmap of image data and inputs that bitmap into a first-in first-out (FIFO) data buffer. A fixed subset of the buffer stored bits forms a sampling window through which a selected block of the bitmap image data may be viewed (for example, a 9×9 block of pixels with the edge pixels truncated). The sampling window contains a center bit cell which changes on each shift of the image bits through the FIFO buffer. As the serialized data is shifted, the sampling window views successive bit patterns formed by pixels located at the window's center bit cell and its surrounding neighbor bit cells. Each bit pattern formed by the center bit and its neighboring bits is compared in a matching network with prestored templates. If a match occurs, indicating that the center bit resides at an image edge and that the pixel it represents can be altered so as to i
Bearss James G.
Bradburn Wayne E.
Jones Arlin R.
Roylance Eugene A.
Mehta Bhavesh M.
Patel Kanji
Simmons Lane R.
LandOfFree
Programmable mapping of lower resolution digital data to a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Programmable mapping of lower resolution digital data to a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Programmable mapping of lower resolution digital data to a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3223872