Browse Prior Art Database

A TECHNIQUE OF NOISE STRIPPING AND CONTOUR ENHANCEMENT FOR RASTER IMAGES

IP.com Disclosure Number: IPCOM000024793D
Original Publication Date: 1982-Feb-28
Included in the Prior Art Database: 2004-Apr-04
Document File: 2 page(s) / 52K

Publishing Venue

Xerox Disclosure Journal

Abstract

This is a process for the elimination of background noise dots on a copy which either existed on the original or were created during the copying process. In a system where the original scan data is converted into binary form, each data bit may be tested by using the surrounding bits as inputs to a predictor, and corrected if there is a high probability that the bit is in error. Thus, in Figure 1, bits X1 through X8 would be the inputs to a predictor. If, for example, all were zeros and Y was a one, Y would probably be an error and would be corrected to a zero. The Figure 2 pattern would work almost as well while using six predictor inputs.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 100% of the total text.

Page 1 of 2

XEROX DISCLOSURE JOURNAL

A TECHNIQUE OF NOISE STRIPPING ANDCONTOURENHANCEMENTFOR U.S. Cl. 358/280 RASTER IMAGES

Proposed Classification

Int. Cl. H04n 1/40

Henry H. Liao

This is a process for the elimination of background noise dots on a copy which either existed on the original or were created during the copying process. In a system where the original scan data is converted into binary form, each data bit may be tested by using the surrounding bits as inputs to a predictor, and corrected if there is a high probability that the bit is in error. Thus, in Figure 1, bits X1 through X8 would be the inputs to a predictor. If, for example, all were zeros and Y was a one, Y would probably be an error and would be corrected to a zero. The Figure 2 pattern would work almost as well while using six predictor inputs.

Larger noise dots may be corrected by using larger matrices of 4, 5, etc., bits on a side. Of course, these require larger amounts of storage for the look-up tables. Rectangular matrices (3x5, 6x3, etc.) may also be used, singly or two in combination, if vertical and/or horizontal corrections are needed.

x1 x2 x3

x4 Y x5

X6 X7 X8

Figure 1

x1 x2 x3

Y

X4 X5 X6

Figure 2

Volume 7 Number 1 January/February 1982 51

[This page contains 1 picture or other non-text object]

Page 2 of 2

    XEROX DISCLOSURE JOURNAL Volume 7 Number 1 January/February 1982

52

[This page contains 1 picture or other non-text object]