Encoding method for deferred anti-aliasing

• US 5,646,751 A
• Filed: 05/25/1995
• Issued: 07/08/1997
• Est. Priority Date: 05/25/1995
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Kellman, Jeffrey N., Motamed, Margaret
• Primary Examiner(s): Coles, Sr., Edward L.
• Assistant Examiner(s): LEE, CHEUKFAN
• Application Number:
  US08/452,625
• Time in Patent Office: 775 Days
• Field of Search: 358/518, 358/523, 358/524, 358/539, 358/534, 358/433, 358/432, 358/445, 358/426, 358/261.2, 358/261.3, 358/261.4, 358/430, 358/533, 358/467, 382/166, 382/167
• US Class Current: 358/518
• CPC Class Codes: H04N 1/644 using a reduced set of representative colours, e.g. each representing a particular range in a colour space

Abstract

A method for increasing the speed of color conversion from a plurality of color spaces to a single color space by reducing the number of color pixels by a factor of two prior to the color conversion step. This pixel reduction constitutes a lossy data compression scheme which takes advantage of the advance knowledge that the pixels will later undergo anti-aliasing by averaging subsequent to the color conversion. First the pixels are grouped into pairs, and one pixel of each pair is eliminated to form a pseudo halfpixel. The eliminated pixel is replaced with a tag identifying either the previous, current or next pseudo halfpixel. Then each pseudo halfpixel is color converted. Finally the indicated previous, current or next pseudo halfpixel, now color converted, is averaged with the current pseudo halfpixel to form a halfpixel, which is then averaged with the halfpixel above or below it to form an anti-aliased full pixel which can be displayed. The color conversion time of this process is roughly halved since half of the pixels are eliminated before the color conversion step and then effectively added back in after the color conversion step.

Claims

1. A method of scanning a picture, the pixels of said picture being in a plurality of input color spaces and a plurality of input pixel densities, to produce a scanline of pixels comprising previous, current and next pixels, and displaying said scanline in a single output color space and a single output pixel density, the processing of each current pixel comprising the steps of:

scanning said picture to generate said scanline of pixels in said plurality of color spaces and said plurality of pixel densities,

replicating and then resampling said pixels to generate a scanline of aligned pairs of subpixels in said single output pixel density,

converting each said aligned pair of subpixels into a pseudo-halfpixel by deleting one subpixel and adding a tag which identifies either the previous, the current, or the next pseudo-halfpixel,

converting said pseudo-halfpixel into the said single output color space,

averaging the converted pseudo-halfpixel with the pseudo-halfpixel identified by said tag to produce a halfpixel, and

displaying said halfpixel.

2. A method of scanning a picture, the pixels of said picture being in a plurality of input color spaces and a plurality of input pixel densities, to produce a scanline of pixels comprising previous, current and next pixels, and displaying said scanline in a single output color space and a single output pixel density, the processing of each current pixel comprising the steps of:

scanning said picture to generate said scanline of pixels in said plurality of color spaces and said plurality of pixel densities,

replicating and then resampling said pixels in two directions to generate two scanlines of subpixels in said single output pixel density, each said scanline consisting of aligned pairs of said subpixels,

converting each aligned pair of subpixels into a pseudo-halfpixel by deleting one subpixel and adding a tag which identifies either the previous, the current, or the next pseudo-halfpixel in the same scanline,

converting said pseudo-halfpixels into said single output color space,

averaging each converted pseudo-halfpixel with the pseudo-halfpixel identified by said tag to produce a halfpixel,

averaging each said halfpixel from the first said scanline with the corresponding said halfpixel from the second said scanline to produce a fullpixel, and

displaying said fullpixel.