Rendering images to lower bits per pixel formats using reduced numbers of registers
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
H04N-001/60
G03F-003/08
H04N-001/46
G06K-009/00
G06K-015/02
출원번호
US-0638743
(2015-03-04)
등록번호
US-9262704
(2016-02-16)
발명자
/ 주소
Metcalfe, David Jon
Metcalfe, Ryan David
출원인 / 주소
Xerox Corporation
대리인 / 주소
Gibb & Riley, LLC
인용정보
피인용 횟수 :
2인용 특허 :
14
초록▼
Methods and systems render higher bit per pixel contone images to lower bit formats using multiple registers of a SIMD processor. The rendering process uses a first register to maintain contone image values of all the pixels being simultaneously processed. A second register maintains a threshold val
Methods and systems render higher bit per pixel contone images to lower bit formats using multiple registers of a SIMD processor. The rendering process uses a first register to maintain contone image values of all the pixels being simultaneously processed. A second register maintains a threshold value used during the conversion process. A third register maintains one value for the print ready format pixels (e.g., those having less bits per pixel), and a fourth register maintains the other value (e.g., 0) for the print ready format pixels. Also, a fifth register maintains the conversion error amount for all the pixels being simultaneously processed. Sixth through ninth registers maintain distributed conversion error amounts produced by the diffusing process (for different pixels being simultaneously processed); and a tenth register maintains the pixels in the print-ready format produced by the conversion for all the pixels being simultaneously processed.
대표청구항▼
1. A method comprising: receiving a contone image having a first amount of bits per pixel into a single-instruction, multiple-data (SIMD) processor;converting said contone image to a print-ready format using said SIMD processor, said print ready format having a second amount of bits per pixel that i
1. A method comprising: receiving a contone image having a first amount of bits per pixel into a single-instruction, multiple-data (SIMD) processor;converting said contone image to a print-ready format using said SIMD processor, said print ready format having a second amount of bits per pixel that is smaller than said first amount of bits per pixel, and said converting producing a conversion error amount for each pixel converted;combining pixels in said print-ready format produced by said converting into a print-ready image; andoutputting said print-ready image,said converting comprising: sequentially converting pixels of said contone image along scanlines of said contone image, said scanlines comprising pixels aligned in a processing direction in said contone image;diffusing a portion of said conversion error amount from a first pixel in a first scanline to at least one other pixel in said first scanline and to at least one other pixel in a second scanline immediately adjacent said first scanline; andsimultaneously converting pixels in immediately adjacent scanlines in a spaced manner that separates pixels being simultaneously processed,said converting simultaneously using different registers of said SIMD processor to maintain each of: pixel values of said contone image; a threshold value used during said converting; said conversion error amount for each pixel; distributed conversion error amounts produced by said diffusing; and said pixels in said print-ready format, in different registers. 2. The method according to claim 1, said scanlines being parallel to one another. 3. The method according to claim 1, said scanlines comprising adjacent pixels forming a straight line in said processing direction in a two-dimensional array of pixels forming said contone image. 4. The method according to claim 1, said spaced manner separates pixels being simultaneously processed by a processing direction pixel distance, said processing direction pixel distance permitting said conversion error amount to be supplied to said pixels being simultaneously processed before simultaneous processing begins on said pixels being simultaneously processed, andsaid processing direction pixel distance being parallel to said processing direction and comprising an amount of pixels aligned in said processing direction. 5. The method according to claim 1, said diffusing providing different portions of said conversion error amount to different pixels. 6. The method according to claim 1, said conversion error amount being based on a difference from a threshold value. 7. The method according to claim 1, said converting only using one scanline of memory to store said conversion error amount for each pixel while simultaneously processing multiple scanlines. 8. A method comprising: receiving a contone image having a first amount of bits per pixel into a single-instruction, multiple-data (SIMD) processor;converting said contone image to a print-ready format using said SIMD processor, said print ready format having a second amount of bits per pixel that is smaller than said first amount of bits per pixel, and said converting producing a conversion error amount for each pixel converted;combining pixels in said print-ready format produced by said converting into a print-ready image; andoutputting said print-ready image,said converting comprising: sequentially converting pixels of said contone image along scanlines of said contone image, said scanlines comprising pixels aligned in a processing direction in said contone image;diffusing a portion of said conversion error amount from a first pixel in a first scanline to at least one other pixel in said first scanline and to at least one other pixel in a second scanline immediately adjacent said first scanline; andsimultaneously converting pixels in immediately adjacent scanlines in a spaced manner that separates pixels being simultaneously processed by a processing direction pixel distance, said processing direction pixel distance permitting said conversion error amount to be supplied to said pixels being simultaneously processed before simultaneous processing begins on said pixels being simultaneously processed,said converting simultaneously using: a first register of said SIMD processor to maintain contone image values having said first amount of bits per pixel for all said pixels being simultaneously processed;a second register of said SIMD processor to maintain a threshold value used during said converting;a third register of said SIMD processor to maintain said conversion error amount for all said pixels being simultaneously processed;a fourth register of said SIMD processor to maintain distributed conversion error amounts produced by said diffusing for all said pixels being simultaneously processed; anda fifth register of said SIMD processor to maintain said pixels in said print-ready format produced by said converting for all said pixels being simultaneously processed. 9. The method according to claim 8, said scanlines being parallel to one another. 10. The method according to claim 8, said scanlines comprising adjacent pixels forming a straight line in said processing direction in a two-dimensional array of pixels forming said contone image. 11. The method according to claim 8, said processing direction pixel distance being parallel to said processing direction and comprising an amount of pixels aligned in said processing direction. 12. The method according to claim 8, said diffusing providing different portions of said conversion error amount to different pixels. 13. The method according to claim 8, said conversion error amount being based on a difference from a threshold value. 14. The method according to claim 8, said converting only using one scanline of memory to store said conversion error amount for each pixel while simultaneously processing multiple scanlines. 15. A method comprising: receiving a contone image having eight bits per pixel into a single-instruction, multiple-data (SIMD) processor;converting said contone image to a print-ready format using said SIMD processor, said print ready format having one bit per pixel, and said converting producing a conversion error amount for each pixel converted;combining pixels in said print-ready format produced by said converting into a print-ready image; andoutputting said print-ready image,said converting comprising: sequentially converting pixels of said contone image along scanlines of said contone image, said scanlines comprising pixels aligned in a processing direction in said contone image;diffusing a portion of said conversion error amount from a first pixel in a first scanline to another pixel in said first scanline and to four pixels in a second scanline immediately adjacent said first scanline; andsimultaneously converting pixels in immediately adjacent scanlines in a spaced manner that separates pixels being simultaneously processed by a four-pixel distance in said processing direction, said four-pixel distance in said processing direction permitting said conversion error amount to be supplied to said pixels being simultaneously processed before simultaneous processing begins on said pixels being simultaneously processed,said converting simultaneously using: a first register of said SIMD processor to maintain contone image values having said eight bits per pixel for all said pixels being simultaneously processed;a second register of said SIMD processor to maintain a threshold value used during said converting;a third register of said SIMD processor to maintain one value of said print ready format having one bit per pixel;a fourth register of said SIMD processor to maintain another value of said print ready format having one bit per pixel;a fifth register of said SIMD processor to maintain said conversion error amount for all said pixels being simultaneously processed;sixth through ninth registers of said SIMD processor to maintain distributed conversion error amounts produced by said diffusing for different pixels being simultaneously processed; anda tenth register of said SIMD processor to maintain said pixels in said print-ready format produced by said converting for all said pixels being simultaneously processed. 16. The method according to claim 15, said scanlines being parallel to one another. 17. The method according to claim 15, said scanlines comprising adjacent pixels forming a straight line in said processing direction in a two-dimensional array of pixels forming said contone image. 18. The method according to claim 15, said four-pixel distance in said processing direction being parallel to said processing direction and comprising pixels aligned in said processing direction. 19. The method according to claim 15, said diffusing providing different portions of said conversion error amount to different pixels. 20. The method according to claim 15, said converting only using one scanline of memory to store said conversion error amount for each pixel while simultaneously processing multiple scanlines.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (14)
Webb Michael (Lane Cove AUX) Brown David R. (East Roseville AUX) Naylor ; Jr. William C. (Mount Kuring-gai AUX), Block parallel error diffusion method and apparatus.
Davies Daniel (Palo Alto CA), Data rotation using parallel to serial units that receive data from memory units and rotation buffer that provides rotat.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.