Method of and apparatus for binarizing image signal

• US 5,530,561 A
• Filed: 03/24/1995
• Issued: 06/25/1996
• Est. Priority Date: 04/15/1994
• Status: Expired due to Term

• Current Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Photo Film Co Limited (Fujifilm Holdings Corporation)
• Inventors: Shimazaki, Osamu
• Primary Examiner(s): Rogers, Scott A.
• Application Number:
  US08/409,605
• Time in Patent Office: 459 Days
• Field of Search: 358/455, 358/456, 358/457, 358/458, 358/465, 358/466, 382/270, 382/252
• US Class Current: 358/3.03
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: H04N 1/4052 by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions
• CPC Class Codes: H04N 1/4053 with threshold modulated relative to input image data or vice versa

Abstract

To obtain a binary gradation image from a continuous gradation image, a multi-valued image signal I(x,y) is compared with a threshold signal TH by a comparator to produce a binary image signal P(x,y), an error signal E(x,y) is determined from the binary image signal P(x,y) and the multi-valued image signal I(x,y), and the error signal E(x,y) is multiplied by an error diffusion coefficient W(k,l) to produce a diffusion error signals ΔE(x,y). A random number R(x,y) generated by a random number generator and a weighting coefficient f(I(x,y)) generated as a function of an image signal by a weighting coefficient generator are multiplied, and the product is added to the diffusion error signals ΔE(x,y). The sum is added to the multi-valued image signal I(x,y) thereby to produce a corrected multi-valued image signal I(x,y), which is then compared with the threshold signal TH by the comparator thereby generating a binary image signal P(x,y).

Claims

1. An apparatus for producing a binary gradation image signal from a continuous gradation image signal, comprising:

a comparing means for comparing a multi-value image signal corresponding to a target pixel with a threshold signal to produce a binary image signal;

an error calculating means for determining an error signal based on a difference between said multi-valued image signal and said binary image signal;

an error diffusion means for determining a set of error diffusion signals based on said error signal and a set of error diffusion coefficients, and applying said error diffusion signals to pixels proximate to said target pixel;

a random number generating means for generating a random number;

a weighting coefficient generating means for generating a weighting coefficient whose value varies depending on a density value of an input multi-valued image signal; and

a weighting means for multiplying said random number by said weighting coefficient;

wherein the random number weighted by said weighting coefficient is applied to at least one of;

said error signal, said error diffusion signals, and said threshold signal.

2. An apparatus according to claim 1, further comprising:

an adding means for adding the weighted random number to said error signal.

3. An apparatus according to claim 1, further comprising:

a product summing means for producing the sum of products, between the error diffusion coefficients and the weighted random number.

4. An apparatus according to claim 1, further comprising:

an adding means for adding the weighted random number to the result of said error diffusion process.

5. An apparatus according to claim 1, further comprising:

an adding means for adding the weighted random number to said threshold signal.

6. A method of producing a binary gradation image signal from a continuous gradation image signal, comprising the steps of:

comparing a multi-valued image signal corresponding to a target pixel with a threshold signal to produce a binary image signal;

determining an error signal based on a difference between said multi-valued image signal and said binary image signal;

determining a set of error diffusion signals based on said error signal and a set of error diffusion coefficients;

applying said error diffusion signals to pixels proximate said target pixel thereby to carry out an error diffusion process;

generating a random number;

generating a weighting coefficient whose value varies depending on a density value of an input multi-valued image signal;

weighting said random number with said weighting coefficient; and

applying the random number weighted by said weighting coefficient to at least one of said error signal, said error diffusion signals, and said threshold signal.

7. A method according to claim 6, further comprising the step of adding the weighted random number to said error signal.

8. A method according to claim 6, further comprising the step of producing the sum of products between the error diffusion coefficients and the weighted random number.

9. A method according to claim 6, further comprising the step of adding the weighted random number to the result of said error diffusion process.

10. A method according to claim 6, further comprising the step of adding the weighted random number to said threshold signal.