Generating imitation custom artwork by simulating brush strokes and enhancing edges

• US 5,844,565 A
• Filed: 09/13/1996
• Issued: 12/01/1998
• Est. Priority Date: 04/15/1994
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Inventors: Mizutani, Masao, Lee, Chuen-Chien
• Primary Examiner(s): Zimmerman, Mark K.
• Application Number:
  US08/713,989
• Time in Patent Office: 809 Days
• Field of Search: 382/276, 382/274, 382/277, 382/282, 382/286, 382/199, 382/263, 382/264, 382/266, 395/433, 395/434, 348/578, 345/429, 345/431, 345/432
• US Class Current: 345/581
• CPC Class Codes: G06T 11/00 2D [Two Dimensional] image generation
• CPC Class Codes: G06T 11/001 Texturing; Colouring; Generation of texture or colour inpainting G06T5/77
• CPC Class Codes: G06T 15/02 Non-photorealistic rendering

Abstract

The precision of an initial image is reduced through image automatic processing in a manner simulating techniques used in the production of a custom art work. The resulting image, although generated without human intervention, has a pleasing, customized appearance which can be desirable to consumers seeking decorative art work. One way of reducing the precision of the initial image is modifying portions of the image to have more uniform luminance, color saturation, or color hue, producing a resulting image appearing to contain brush strokes. Alternatively, the precision of the initial image is reduced by locating contrast edges in the image, and modifying the luminance along these edges, producing a resulting image in which objects and features are outlined and have an enhanced definition from the background.

Claims

1. A method for modifying an initial image to produce a resulting image having the appearance of custom art work, comprising

storing a pixellated representation of said initial image in an electronic memory,

inspecting pixels of said pixellated representation of said initial image to derive, without human input, modifications which reduce the precision of said initial image in a manner simulating techniques used by humans to produce custom art work,

automatically applying said modifications to said pixels of said pixellated representation of said initial image to generate a pixellated representation of said resulting image, by blurring said initial image to form a background, and then selecting and modifying portions of said initial image over said blurred background, and

producing a visual output of said resulting image.

2. The method of claim 1 wherein producing a visual output of said resulting image comprises one or more of

printing a copy of said pixellated representation of said resulting image and

generating a display of said pixellated representation of said resulting image on an electronic display device.

3. The method of claim 1 wherein modifying a selected portion comprises

selecting a luminance and color value corresponding to one or more pixels within said selected portion of said initial image,

comparing said selected luminance value to an average luminance value derived from a plurality of pixels of said initial image,

reducing said selected luminance value if said selected luminance value is less than said average luminance value, otherwise increasing said selected luminance value,

deriving pixel luminance and color values from said selected color value and said reduced or increased selected luminance value, and

modifying pixels of said selected portion to approach said derived pixel luminance and color values.

4. The method of claim 3 wherein deriving pixel luminance and color values comprises augmenting said selected color value and said reduced or increased selected luminance value with pseudo-random noise to produce variation in luminance or color of pixels of said selected portion.

5. The method of claim 1 wherein selecting portions of said initial image comprises selecting elongated portions having substantially parabolic curved elongated edges.

6. The method of claim 1 wherein selecting portions of said initial image further comprises selecting elongated portions oriented in one of at least two possible directions.

7. The method of claim 1 wherein

selecting portions of said initial image comprises selecting a first set of rows of blocks extending through said image, and then selecting a second set of rows of blocks interleaved with said first set of rows of blocks and extending through said image, and

modifying said portions comprises modifying portions along said first set of rows prior to modifying portions along said second set of rows,

whereby modifications of said portions are variably overlapped within said image.

8. The method of claim 1 wherein

selecting portions of said initial image comprises selecting large portions and smaller portions of said image, and

modifying said portions comprises modifying said large portions prior to modifying said smaller portions.

9. The method of claim 8 wherein

selecting large portions of said image comprises selecting large portions along a first set of rows extending through said image, and selecting large portions along a second set of rows interleaved with said first set of rows and extending through said image, and

modifying said portions comprises modifying portions along said first set of rows prior to modifying portions along said second set of rows,

whereby modifications of said large portions are variably overlapped within said image.

10. The method of claim 8 wherein selecting smaller portions of said image comprises determining contrast of said image and selecting smaller portions in regions of said image having high contrast.

11. The method of claim 1 wherein selecting and modifying portions of said pixellated representation of said initial image to reduce the precision of said initial image comprises

locating contrast edges in the initial image, and

modifying luminance of said initial image along said edges, whereby objects in said resulting image have enhanced definition from background.

12. The method of claim 11 wherein modifying luminance of said initial image along said edges comprises increasing luminance of said image along edges in relatively high luminance areas of said image, and decreasing luminance of said image along edges in relatively low luminance areas of said image.

13. The method of claim 11 wherein modifying luminance of said initial image along said edges comprises

generating at each pixel of said initial image a differential measure of luminance contrast in two directions, and

modifying luminance of pixels of said initial image by combining each said differential measure with luminance of a corresponding pixel in said initial image, to generate luminance values for said resulting image.

14. The method of claim 13 wherein generating a differential measure of contrast at a pixel includes computing a luminance variation between pixels neighboring said pixel.

15. The method of claim 13 wherein combining each said differential measure with its corresponding pixel in said initial image comprises multiplying each said differential measure with luminance of a corresponding pixel in said initial image.

16. The method of claim 13 wherein combining each said differential measure with a corresponding pixel in said initial image comprises

determining whether said differential measure is below a threshold value, and if so modifying said differential measure by subtracting it from a predetermined maximum value, and

multiplying each thus modified differential measure with luminance of said corresponding pixel in said initial image.

17. The method of claim 13 wherein combining each said differential measure with a corresponding pixel in said initial image comprises

deriving an average luminance value for said initial image from a plurality of pixels of said initial image,

determining whether luminance of said corresponding pixel in said initial image is above said average luminance value, and if so modifying said differential measure by subtracting it from a predetermined maximum value, and

multiplying each thus modified differential measure with luminance of said corresponding pixel in said initial image.

18. Apparatus for modifying an initial image to produce a resulting image having the appearance of custom art work, comprising

an electronic memory storing a pixellated representation of said initial image,

computing circuitry inspecting pixels of said pixellated representation of said initial image to derive, without human input, modifications which reduce the precision of said initial image in a manner simulating techniques used by humans to produce custom art work, and automatically applying said modifications to said pixels of said pixellated representation of said initial image to generate a pixellated representation of said resulting image, by blurring said initial image to form a background, and then selecting and modifying portions of said initial image over said blurred background, and

an output port for connection to an output device capable of producing a visual output of said resulting image.

19. The apparatus of claim 18 wherein said output port is configured for connection to a printer.

20. The apparatus of claim 18 wherein said output port is configured for connection to an electronic display monitor.

21. The apparatus of claim 18 wherein

said computing circuit selects portions of said initial image by selecting a first set of rows of blocks extending through said image, and then selecting a second set of rows of blocks interleaved with said first set of rows of blocks and extending through said image, and

said computing circuit modifies portions along said first set of rows prior to modifying portions along said second set of rows,

whereby modifications of said portions are variably overlapped within said image.

22. The apparatus of claim 18 wherein

said computing circuit selects large portions and smaller portions of said image, and

said computing circuit modifies said large portions prior to modifying said smaller portions.

23. The apparatus of claim 22 wherein said computing circuit selects smaller portions of said image by determining contrast of said image and selecting smaller portions in regions of said image having high contrast.

24. The apparatus of claim 18 wherein said computing circuitry modifies pixels of said pixellated representation of said initial image to reduce the precision of said initial image by locating contrast edges in the initial image, and modifying luminance of said initial image along said edges, whereby objects in said resulting image have enhanced definition from background.

25. The apparatus of claim 24 wherein said computing circuitry modifies luminance of said initial image along said edges by increasing luminance of said image along edges in relatively high luminance areas of said image, and decreasing luminance of said image along edges in relatively low luminance areas of said image.

26. The apparatus of claim 24 wherein said computing circuitry modifies luminance of said initial image along said edges by generating at each pixel of said initial image a differential measure of luminance contrast in two directions, and modifying luminance of pixels of said initial image by combining each said differential measure with luminance of a corresponding pixel in said initial image, to generate luminance values for said resulting image.

27. The apparatus of claim 26 wherein said computing circuitry generates a differential measure of contrast at a pixel by computing a luminance variation between pixels neighboring said pixel.

28. The apparatus of claim 26 wherein said computing circuitry combines each said differential measure with its corresponding pixel in said initial image by multiplying each said differential measure with luminance of a corresponding pixel in said initial image.