Edge enhancement error diffusion thresholding for document images

• US 5,034,990 A
• Filed: 05/08/1990
• Issued: 07/23/1991
• Est. Priority Date: 09/05/1989
• Status: Expired due to Term

• Current Assignee: Eastman Kodak Company
• Original Assignee: Eastman Kodak Company
• Inventors: Klees, Kevin J.
• Primary Examiner(s): NOT, DEFINED
• Assistant Examiner(s): Santos, Daniel
• Application Number:
  US07/520,587
• Time in Patent Office: 441 Days
• Field of Search: 382/50, 382/54, 382/22, 358/447, 358/462, 358/464, 358/465
• US Class Current: 382/252
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: H04N 1/4053 with threshold modulated relative to input image data or vice versa

Abstract

Apparatus for converting multilevel video images into a bilevel image having an accurate rendition of the original grayscale of the image. The apparatus performs enhanced preservation of edge information in the image when using error diffusion techniques.

Claims

1. Apparatus for performing error diffusion thresholding of multiple level digital images into bilevel digital images using image edge information to select from several different lookup tables to perform the calculation of an error signal to be diffused, said apparatus comprising:

means for scanning a grayscale image and producing data output representative of the density levels of individual pixels comprising that image;

means for defining a matrix of pixels;

means for delaying input pixel data by a predetermined number of scan lines;

means for detecting that portion of the image being scanned to determine if said input pixel data is text or continuous tone data and generate a signal accordingly;

means for generating multiple error values from first and second sets of error values resulting from first and second predefined error functions when said multiple error values contain either positive or negative signs;

means for selecting in response to said generated signal either said first or second sets of error values from a preselected set of values containing either positive or negative sign information;

means for generating a signal representing the edge information at the present scan location in the image;

means for selecting one of said multiple error values in said selected set based on said edge signals;

means for adding or subtracting said selected error values based on said edge signal;

means for adding or subtracting said selected error value from said predetermined number of delayed pixel values; and

means for generating an output value.

2. The apparatus in claim 1 wherein said adding and subtracting means further installs means for checking to insure that the calculated value is within a predetermined range.

3. The apparatus in claim 1 wherein said first and second sets of error values resulting from first and second predefined error functions are loaded into said lookup tables by an external source.

4. The apparatus in claim 1 wherein edge signal generating means is a subtraction circuit.

5. The apparatus in claim 1 wherein edge signal generating means is a programmable read only memory.

6. A method for performing error diffusion thresholding of multiple level digital images into bilevel digital images using image edge information to select from several different lookup tables to perform the calculation of an error signal to be diffused, said method comprising the following steps:

scanning a grayscale image and producing a data output representative of the density levels of individual pixels comprising that image;

delaying the input pixel data by a predetermined number of scan lines;

detecting that portion of the image being scanned to determine if the input pixel is text or continuous tone data and generate a signal corresponding thereto;

generate first and second sets of error values resulting from first and second predefined error functions loaded into said lookup table;

select in response to said generated signal either said first or said second set of error values stored in said lookup tables;

generate a signal representing the edge information at the present scan location in the image;

selecting an error value from said selected set based on said signal representing the edge information;

adding or subtracting said selected error values based on said edge signal;

adding or subtracting said selected error value from the predetermined number of delayed pixel values; and

generate an output value.