Mixed screen frequencies with image segmentation

• US 5,471,543 A
• Filed: 04/30/1992
• Issued: 11/28/1995
• Est. Priority Date: 06/25/1990
• Status: Expired due to Fees

• Current Assignee: NEXPRESS SOLUTIONS LLC
• Original Assignee: Eastman Kodak Company
• Inventors: Tai, Hwai-Tzuu, Ng, Yee S.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US07/877,353
• Time in Patent Office: 1,307 Days
• Field of Search: 358/453, 358/456, 358/457, 358/459, 358/462, 358/464, 358/298, 382/9, 382/50
• US Class Current: 382/173
• CPC Class Codes: H04N 1/4058 with details for producing a halftone screen at an oblique angle H04N1/4056 takes precedence

Abstract

Apparatus for producing a screen halftone image provides for segmentation of images by optical density, busyness, segment size, and/or propensity to contouring; and for printing the segments which have high optical density, are more busy, and/or have a lesser propensity to contouring at a relatively high screen frequency; and for printing segments which have low optical density, are less busy, and/or have a greater propensity to contouring at a relatively low screen frequency. Both screen frequencies are printed at the same apparent screen angle to reduce texturing. Small segments are assigned the same screen frequency as the surrounding segment.

Claims

1. Apparatus for producing a screened halftone image of continuous tone original images, said apparatus comprising:

means for producing a series of input signals whose values are representative of the gray levels of successive picture elements of an original image;

means for examining each pixel for an image characteristic and (1) means for classifying pixels by said image characteristic and (2) means for grouping pixels having the same classification to form segmented regions; and

means for printing groups of sub-elements sequentially organized into two dimensional matrix halftone cells, each halftone cell being associated with one of said input signals;

said printing means being adapted to print halftone cells associated with one of said segmented regions at a first screen frequency and to print halftone cells associated with another of said segmented regions at a second screen frequency higher than said first screen frequency.

2. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment the original image into regions of different optical densities.

3. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment the original image into regions of different optical densities such that the optical density of said one segmented region is lower than the optical density of said other segmented region.

4. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment the original image into regions of different image busyness.

5. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment busyness such that said one segmented region is less busy than said other segmented region.

6. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment the original image into regions of different image propensity to countouring.

7. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to segment the original image into regions of different image propensity to countouring such that said one segmented region includes image regions having a greater propensity to contouring than image regions included in said other segmented region.

8. Apparatus as set forth in claim 1 wherein said segmenting means is adapted to include segments below a predetermined size in the surrounding segment.

9. Apparatus for producing a screened halftone image of continuous-tone original images, said apparatus comprising:

means for producing a series of input signals whose values are representative of the gray levels of successive picture elements of an original image;

means for segmenting the original image into at least two regions of different image characteristics, said segmenting means including means for examining each pixel for an image characteristic and (1) means for classifying pixels by said image characteristic and (2) means for grouping pixels having the same classification to form a region;

means for printing groups of sub-elements sequentially organized into two dimensional matrix halftone cells which have substantially the same apparent screen angle, each halftone cell being associated with one of said input signals;

said printing means being adapted to print halftone cells associated with one of said segmented regions at a first screen frequency and to print halftone cells associated with the other of said segmented regions at a second screen frequency higher than said first screen frequency.

10. Apparatus as set forth in claim 9 wherein said segmenting means is adapted to segment the original image into regions of different optical densities.

11. Apparatus as set forth in claim 9 wherein said segmenting means is adapted to segment the original image into regions of different optical densities such that the optical density of said one segmented region is lower than the optical density of said other segmented region.

12. Apparatus as set forth in claim 9 wherein said segmenting means is adapted to segment the original image into regions of different image busyness.

13. Apparatus as set forth in claim 9 wherein said segmenting means is adapted to segment the original image into regions of different image propensity to countouring.

14. Apparatus as set forth in claim 9 wherein said segmenting means is adapted to include segments below a predetermined size in the surrounding segment.

15. A method for producing a screened halftone image of continuous-tone original images, said method comprising the steps of:

generating a series of input signals whose values are representative of the grey levels of successive picture elements of an original image;

segmenting the original image into at least two regions of different image characteristics, said step of segmenting including examining each pixel for an image characteristic and (1) classifying pixels by said image characteristic and (2) grouping pixels having the same classification to form a region;

printing groups of sub-elements sequentially organized into two dimensional matrix halftone cells, each halftone cell being associated with one of said input signals and wherein halftone cells associated with one of said segmented regions are printed at a first screen frequency and halftone cells associated with the other of said segmented regions are printed at a second screen frequency higher than said first screen frequency.

16. The method of claim 15 and wherein in said segmentation step, the grey level of each picture element and the grey levels of picture elements contiguous with said each picture element are used to calculate an image gradient strength for said each picture element, each picture element is classified in accordance with a category of image gradient strength and groups of contiguous pixels of similar gradient strengths are grouped together into distinct regions which have their own identity codes.

17. The method of claim 16 and wherein a region is examined to determine if it is a busy region and the size of the region is examined and its average intensity compared with a threshold and in response to the examination for being busy and size and average intensity a decision is made to print the region with a predetermined screen frequency.

18. The method of claim 16 and wherein small regions are merged into surrounding larger regions and each resulting region is evaluated to determine a screen frequency for recording the pixels in the region.

19. The method of claim 17 and wherein a region classified as not busy is rendered in a lower frequency screen pattern.

20. The method of claim 15 and wherein the halftone cells each have substantially the same screen angle.

21. The method of claim 16 and wherein the halftone cells each have substantially the same screen angle.

22. The method of claim 15 and wherein the grey level value of each pixel of the original image is examined and determined to be within an identified brightness range if between two threshold values and wherein pixels within the same identified brightness range are clustered into the same region to segment the image.

23. The method of claim 22 and wherein edge detection is used to determine an outline in an image during the segmenting step.

24. The method of claim 15 and wherein a property of each pixel of the original image is examined and neighboring pixels that possess a similar property are grouped with said each pixel to grow a region.