Image analysis – Pattern recognition – Template matching
Reexamination Certificate
2000-05-25
2004-08-10
Mariam, Daniel (Department: 2621)
Image analysis
Pattern recognition
Template matching
C382S218000, C382S266000, C358S447000
Reexamination Certificate
active
06775410
ABSTRACT:
BACKGROUND OF THE INVENTION AND MATERIAL DISCLOSURE STATEMENT
The present invention relates to a methodology for improving the print quality of line-art corners and other fine details as found in both font and image data. Due to the nature of inks, paper and printing, these types of problems are well known in the printing arts.
In the printing arts this problem was typically overcome by manual image customization by hand of any troublesome detail areas. This was particularly the case with fonts or reusable type. For example, ink traps would be added to those areas in a font character where too much ink would otherwise collect and cause smearing. Similarly, detail areas would be sharpened to insure the desired print result.
This approach of compensating to get a desired result such as sharpening has followed-on from the print arts into the digital imaging arts. As an example,
Digital Typography: An Introduction to Type and Composition for Computer System Design,
by Richard Rubinstein, discusses the desirability of compensating for electrostatic effects which result in toner not being placed on the paper exactly as the bit image specifies. Compensation is depicted there as adding to the image bit map to sharpen convex (outside) corners which would otherwise get rounded over. An alternative compensation is also depicted for handling situations involving concave (inside) corners by removing black printing pixels from the corner region of a shape to make an ink trap. In
FIG. 18
there is depicted the bit image data and thereby the desired print image. In
FIG. 19
is depicted the rounded-over result when printed on a given printer.
FIG. 20
shows the bit image compensation which when printed will yield the desired result as depicted in
FIG. 18
above. Note the ears
2000
added to the outside corners and the ink trap
2010
added to address the inside corner.
The following appear to be relevant to manipulation and enhancement of the edges of image shapes. They are incorporated by reference in their entirety for their teaching, particularly of template matching, and of bit map or signal substitution.
In U.S. Pat. No. 4,847,641 to Tung, print enhancement circuitry to enhance the printed image produced by a laser beam printer is interposed between the character generator circuits and the laser drive circuits to modify the laser drive signals provided by the character generator circuits. Bit data representing successive lines of the bit map for a desired image are stored in a first-in first-out (FIFO) buffer. The bit pattern sample window having a central cell (bit) and a selected (arbitrary) number of neighboring bits is compared to a number of matching bit patterns or templates, each of which is associated with an error element or cell. When a logic matching network detects a match, a modification signal associated with a unique compensation cell (bit) is generated. The sample window central bit is then replaced (modified) with the unique compensation bit required by the matching template. In this manner, all bits in a desired bit map, or set of bit maps, are examined and their corresponding laser drive signals modified to compensate for the errors associated with the matched templates in a piece-wise manner.
In U.S. Pat. No. 5,383,036 to Mailloux et al., a method for enhancing the contour fidelity of printed images of two or more colors is described, which includes obtaining a digital representation of the color image and finding color separations of each color. Each color separation is enhanced by a single set of inverse symmetrical templates, the set including templates in which the second template is always the inverse of the first, and the third and fourth templates are 180 degree rotations of the first two. The resulting smoothed color separations are recombined into an enhanced image without separation error.
In U.S. Pat. No. 4,437,122 to Walsh et al., presents an invention which enhances the resolution and quality of characters of a system receiving the information initially in the form of video display pixels and providing hard copy output. This is accomplished by storing at least three successive lines of video data in successive, parallel connected shift registers, applying the outputs of the shift registers to a decoder, and generating driving signals for the printer head. The decoder compares the pixels on the same line as well as in preceeding and succeeding lines that surround each specific input pixel to generate the printer head driving signals according to whether straight or curved line segments are to be formed. In effect, the printer increases the density of the information elements and simultaneously provides rounding off of character edges and smoothing of diagonals.
Therefore, there exists a need for a technique which will solve the problem of corner rounding in an automated, non-manual processing operation. Thus, it would be desirable to solve this and other deficiencies and disadvantages with an automated processing methodology.
SUMMARY OF THE INVENTION
The present invention relates to a method of corner sharpening in the display of a bitmapped digital image. The method comprises identifying pixel locations near a corner by comparing the bitmap digital image to templates and, substituting pixel values in the identified pixel locations to achieve a clustering of the substituted pixel values about the corner.
More particularly, the present invention relates to a method of corner sharpening in the display of a bitmapped digital image, by identifying pixel locations near a corner by comparing the bitmap digital image to templates and, generating a signal appropriate to modify the writing of the bitmap digital image and achieve a clustering of the substituted pixel values about the corner.
REFERENCES:
patent: 4389677 (1983-06-01), Rushby et al.
patent: 4437122 (1984-03-01), Walsh et al.
patent: 4847641 (1989-07-01), Tung
patent: 5237646 (1993-08-01), Bunce
patent: 5383036 (1995-01-01), Mailloux et al.
patent: 5987221 (1999-11-01), Bears et al.
patent: 6021256 (2000-02-01), Ng et al.
patent: 6195467 (2001-02-01), Asimopoulos et al.
patent: 6430321 (2002-08-01), Choo
patent: 6504947 (2003-01-01), Nozaki et al.
Kubota, et al. “Edge based probabilistic relaxation for sub-pixel contour extraction”, Department of Computer Science and Engineering, University of South Carolina, pp. 1-16, 1997.*
Nitzberg, et al. “Nonlinear image filtering with edge and corner enhancement”, IEEE, pp. 826-833, 1992.*
Richard Rubinstein,Digital Typography: An Introduction to Type and Compostion for Computer System Design, Addison-Wesley Publishing Company, 1988, pp. 78-81.
Robert P. Loce, Clara Cuciurean-Zapan and Leon Williams, Serial No. 09/505,875, “Loose-Gray-Scale Template Matching”.
R. Loce, M. Cianciosi, and Ying Wei Lin, Ser. No. 09/219,705, “Systems and Methods for Template Matching of Multicolored Images”.
Cuciurean-Zapan Clara
Loce Robert P.
Zhu Wei
Mariam Daniel
Wait Christopher D.
Xerox Corporation
LandOfFree
Image processing method for sharpening corners of text and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing method for sharpening corners of text and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method for sharpening corners of text and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3293843