Image processing method and apparatus

• US 6,118,547 A
• Filed: 07/17/1997
• Issued: 09/12/2000
• Est. Priority Date: 07/17/1996
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Tanioka, Hiroshi
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US08/896,098
• Time in Patent Office: 1,153 Days
• Field of Search: 382/237, 382/252, 382/270, 358/534, 358/456, 358/1.9, 358/535, 358/536, 358/457, 358/458, 358/466, 358/298, 358/460, 395/109
• US Class Current: 358/1.9
• CPC Class Codes: H04N 1/40087 Multi-toning, i.e. converting a continuous-tone signal for reproduction with more than two discrete brightnesses or optical densities, e.g. dots of grey and black inks on white paper
• CPC Class Codes: H04N 1/4052 by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions

Abstract

Provided is an image processing apparatus in which high-quality image representation is possible at very low cost even if high image resolution is obtained. The apparatus includes a first halftone processor for pseudo-halftoning an 8-bit image signal to a 4-bit image signal, an image processor, which comprises a color correcting unit, a converter, a resolution converter, a spatial filter and a gamma corrector, for processing the 4-bit image signal, a second halftone processor for pseudo-halftoning the image-processed 4-bit image signal from the image processor to one bit, and a recorder for outputting an image based upon the 1-bit signal. In another aspect, the amount of calculation required for highly precise color correction processing inclusive of image-quality adjustment processing is large. This has made it necessary to rely upon hardware such as an ASIC, a consequence of which is higher cost. Accordingly, a first halftone processor converts input 8-bit data to 4-bit data and a color correcting unit performs color correction processing, using a LUT conversion, of the 4-bit data. Since the LUT is updated in dependence upon adjustment of image quality performed using a control panel, it is possible for the adjustment to be reflected in real time.

Claims

1. An image processing apparatus for processing a multivalued image signal, comprising:

first halftone processing means for applying pseudo-halftone processing to an L-bit image signal to thereby generate an M-bit image signal, where M<

L holds, said first halftone processing means having means for generating a quotient and a remainder obtained by dividing input data by a predetermined value and said first halftone processing means having means for generating lower order bits based upon a comparison between the remainder and a random-number value;

image processing means for executing image-signal multivalued processing using the M-bit image signal; and

second halftone processing means for generating, from the multivalued-processed M-bit image signal, an image signal of predetermined bits corresponding to output means, wherein said first and second halftone processing means perform halftone processing in a density preserving type quantizing method.

2. The apparatus according to claim 1, wherein said image processing means includes:

color correcting means for performing a color correction such as black generation by processing;

zooming means having an image delay memory; and

spatial filter means.

3. The apparatus according to claim 2, wherein said first halftone processing means generates a 4-bit image signal from an 8-bit image signal, and said color correcting means is a LUT having a 12-bit input.

6. The apparatus according to claim 1, wherein said second halftone processing means includes:

means for delaying and holding an error generated when a conversion to S bits is made;

means for obtaining a weighted mean value using conversion errors and whole-number coefficients of a plurality of pixel positions neighboring a pixel of interest;

error correcting means for correcting an input image signal of the pixel of interest by a value obtained by dividing the weighted mean value by the sum total of the coefficients;

converting means for converting this corrected value to an S-bit signal; and

means for calculating an error generated by the conversion.

7. The apparatus according to claim 6, wherein said error correcting means corrects an error, which is generated by division, when a neighboring pixel is converted.

8. The apparatus according to claim 6, wherein said error calculating means corrects an error, which is generated by rounding of bits, when a neighboring pixel is converted.

9. The apparatus according to claim 6, wherein said error calculating means includes means for reducing generated error to less than a predetermined value and corrects an error, which is generated by reduction processing, when a neighboring pixel is converted.

10. The apparatus according to claim 1, further comprising updating means for updating an image processing method of said image processing means based upon designation of image-quality adjustment.

11. The apparatus according to claim 10, wherein said image processing means includes at least one LUT, and said updating means updates the content of said LUT.

12. The apparatus according to claim 11, wherein said LUT is a LUT for color correction.

4. An image processing apparatus for processing a multivalued image signal, comprising:

first halftone processing means for applying pseudo-halftone processing to an L-bit image signal to thereby generate an M-bit image signal, where M<

L holds, said first halftone processing means has means for generating a quotient and a remainder obtained by dividing input data by a predetermined value and said first halftone processing means having means for generating lower order bits based upon a comparison between the remainder and a random-number value;

means for creating output data from the quotient and lower order bits;

image processing means for executing image-signal multivalued processing using the M-bit image signal

wherein said image processing means includes;

color correcting means for performing a color correction such as black generation by processing;

zooming means having an image delay memory; and

spatial filter means; and

second halftone processing means for generating, from the multivalued-processed M-bit image signal, an image signal of predetermined bits corresponding to output means,

wherein said first and second halftone processing means perform halftone processing in a density preserving type quantizing method.

5. The apparatus according to claim 4, wherein at least one of said means is a LUT.

13. An image processing method for processing a multivalued image signal, comprising the steps of:

reducing number of bits from L bits to M (<

L) bits by pseudo-halftone processing prior to processing of an L-bit image signal by dispersing a remainder to lower order bits of a quotient obtained by dividing input data by a predetermined number of bits;

performing subsequent multivalued processing of image signal using the M-bit image signal; and

generating, from the multivalued-processed M-bit image signal, an image signal of predetermined bits corresponding to output means,

wherein, in said reducing step and generating step, halftone processing in a density preserving type quantizing method is performed.

14. The method according to claim 13, wherein L is 8 and M is 4, and said predetermined number is 17.

15. An image processing method for processing a multivalued image signal, comprising the steps of:

reducing number of bits from L bits to M (<

L) bits by pseudo-halftone processing prior to processing of an L-bit image signal by diffusing a remainder to lower order bits of a quotient obtained by dividing input data by a predetermined number of bits wherein diffusion of the remainder includes comparing the remainder and a random number and varying the least significant bit of the quotient based upon the results of the comparison;

performing subsequent multivalued processing of image signal using the M-bit image signal; and

generating, from the multivalued-processed M-bit image signal, an image signal of predetermined bits corresponding to output means,

wherein, in said reducing step and generating step, halftone processing in a density preserving type quantizing method is performed.

16. A product comprising a processor readable medium including a program for processing multilevel image data, said program including at least:

first program code means for controlling reduction of number of bits from L bits to M (<

L) bits by pseudo-halftone processing prior to processing of an L-bit image signal the first program code means generating a quotient and a remainder obtained by dividing input data by a predetermined value, the first program code means generating lower order bits based upon a comparison between the remainder and a random-number value;

second program code means for controlling multivalued processing of a subsequent image signal using the M-bit image signal; and

third program code means for controlling pseudo-halftone processing from the multivalued-processed image signal to predetermined bits corresponding to output means,

wherein said first and third program code means perform halftone processing in a density preserving type quantizing method.

17. A product according to claim 16, wherein said processor readable medium further includes image data processed or to be processed.