Video signal processing with added probabilistic dither

• US 5,404,427 A
• Filed: 12/24/1992
• Issued: 04/04/1995
• Est. Priority Date: 12/04/1986
• Status: Expired due to Term

• Current Assignee: Quantel Limited (Belden Inc.)
• Original Assignee: Quantel Limited (Belden Inc.)
• Inventors: Long, Robert, Stapleton, Alan L., Cawley, Robin A., Hinson, Neil R.
• Primary Examiner(s): Herndon, Heather R.
• Assistant Examiner(s): SMITH, MICHAEL
• Application Number:
  US07/996,698
• Time in Patent Office: 831 Days
• Field of Search: 395/131, 395/128, 395/129, 358/80, 340/726, 340/735, 345/153
• US Class Current: 345/597
• CPC Class Codes: G06F 3/04845 for image manipulation, e.g. dragging, rotation, expansion or change of colour
• CPC Class Codes: H04N 1/46 Colour picture communication systems
• CPC Class Codes: H04N 9/74 for obtaining special effects H04N9/65 - H04N9/73 take precedence

Abstract

A system in which a characteristic of individual picture points is provided to an accuracy of m binary bits but conveyed by n bits, where n is less than m. The value of a lower order bit of the n bit signal is switched to cause said binary value to represent, for any one picture point, either a value above or a value below the original value. The new values are distributed without order among the picture points with a probability dependent upon the value of the (m-n) lowest order bits of the desired value.

Claims

1. A video signal processing system comprising:

a first defining circuit for defining an image characteristic, said characteristic having a plurality of different values at discrete intervals;

a second defining circuit for defining a desired image characteristic value, for each of a plurality of image elements, within the discrete intervals defined by said first defining circuit;

a determining circuit for determining, for each element of said plurality of image elements, an associated number related to a difference in value between the desired image characteristic value defined by said second defining circuit and a value of said image characteristic defined by said first defining circuit;

a random number generator for generating a random number;

a comparator for comparing the random number and the associated number in order to determine a relationship between the random number and the associated number; and

an output circuit for outputting a value of said image characteristic defined by said first defining circuit depending on the relationship determined by said comparator, the value output by the output circuit representing the desired value defined by said second defining circuit.

2. A video signal processing method comprising:

(a) defining an image characteristic, said characteristic having a plurality of available different values at discrete intervals;

(b) defining a desired image characteristic value, for each of a plurality of image elements, within the discrete intervals;

(c) determining, for each element of said plurality of image elements, an associated number related to a difference in value between the desired characteristic value and an available value of said image characteristic;

(d) generating a random number;

(e) comparing the random number and the associated number in order to determine a relationship between the random number and the associated number; and

(f) outputting a value of said image characteristic from said available values depending on the relationship determined by the comparison, the outputted available value representing the desired value.

3. An electronic processing apparatus for processing digital image data defining pixels which together represent at least one image, wherein output image data representing each pixel in terms of one of a multiplicity of available characteristic values is produced from input image data representing each pixel in terms of a desired characteristic value which is defined to an accuracy greater than that of the output data such that the desired characteristic value can be defined as a value between adjacent available characteristic values, the apparatus comprising:

a source of input image data defining the desired characteristic value of each pixel in terms of a high order number corresponding to one of said multiplicity of available values and a low order number representing the difference between the desired value and said corresponding available value;

a random number generator for generating data representing a random number;

a comparator for comparing the random number and the low order number and for producing a binary value depending on the relationship between the random number and the low order number; and

an output circuit for outputting as the output image data an available characteristic value selected depending on the value of the high order number and the binary value.

4. An apparatus as claimed in claim 3, wherein said input image data for each pixel comprises an m-bit binary number representing said desired characteristic value, with the most significant n-bits of said m-bit binary number corresponding to said high order number and the least significant m-n bits of said m-bit binary number corresponding to said low order number, where m and n are integers and m-n is an integer greater than one.

5. An apparatus as claimed in claim 3, wherein said source of input image data comprises a drawing processor.

6. An apparatus as claimed in claim 3, wherein said source of input image data comprises a frame store.

7. An apparatus as claimed in claim 3, wherein said output circuit comprises an adder for incrementing the corresponding available characteristic value depending on the binary value.

8. An apparatus as claimed in claim 3, wherein said image comprises a plurality of image lines, said source of input image data comprises a computer for providing for each pixel a line count number corresponding to said low order number and indicating the line in which said pixel is positioned, and for providing data corresponding to said high order number representing a first available characteristic value and a second available characteristic value adjacent said first available characteristic value.

9. An apparatus as claimed in claim 8, wherein said output circuit comprises a switch for switching between said first available characteristic value and said second available characteristic value.

10. An apparatus as claimed in claim 3, wherein:

said source of input image data comprises a first multiplier for multiplying image data for a first image by a first weighting factor, a second multiplier for multiplying image data for a second image by a second weighting factor, said first and second multipliers respectively outputting first and second intermediate data comprising an intermediate high order number and an intermediate low order number, a first adder for adding said first and second intermediate low order numbers to produce a low order sum and a carry which together correspond to said low order number, a second adder for adding said first and second intermediate high order numbers and said carry to produce a high order sum corresponding to said high order number;

said random number generator and comparator comprise a look-up table responsive to the low order sum from the first adder for generating said binary value; and

said output circuit comprises a third adder for incrementing the high order sum output from the second adder depending on the binary value to produce the output image data.

11. An apparatus as claimed in claim 10, wherein said random number generator and said comparator together comprise an addressable memory containing data representing multiple look-up tables stored at different respective addresses in the memory and each table containing a set of binary data distributed randomly in the table with a probability related to the address at which the table is stored, and a counter for generating address data for selecting a look-up table in the memory, and wherein said low order number is applied to the memory so as to select a binary datum from an addressed look-up table as said binary value.

12. An apparatus as claimed in claim 3, wherein said image characteristic value comprises the color of each pixel.

13. A method of processing digital image data defining pixels which together represent at least one image, wherein output image data representing each pixel in terms of one of a multiplicity of available characteristic values is produced from input image data representing each pixel in terms of a desired characteristic value which is defined to an accuracy greater than that of the output data such that the desired characteristic value can be defined as a value between available characteristic values, the method comprising:

obtaining input image data defining the desired characteristic value of each pixel in terms of a high order number corresponding to one of said multiplicity of available values and a low order number representing the difference between the desired value and said corresponding available value;

generating data representing a random number;

comparing the random number and the low order number and producing a binary value depending the relationship between the random number and the low order number; and

outputting as the output image data an available characteristic value selected depending on the value of the high order number and the binary value.

14. A method as claimed in claim 13, wherein said input image data for each pixel comprises an m-bit binary number representing said desired characteristic value, with the most significant n-bits of said m-bit binary number corresponding to said high order number and the least significant m-n bits of said binary number corresponding to said low order number, where m and n are integers and m-n is an integer greater than one.

15. A method as claimed in claim 13, wherein said outputting of output image data comprises incrementing the corresponding available characteristic value depending on the binary value.

16. A method as claimed in claim 13, wherein said image comprises a plurality of image lines, the method comprising the further steps of:

providing for each pixel a line count number indicating the line in which said pixel is positioned, which line count number comprises said low order number; and

providing data corresponding to said high order number representing a first available characteristic value and a second available characteristic value.

17. A method as claimed in claim 16, wherein said outputting comprises switching between said first available characteristic value and said second available characteristic value.

18. A method as claimed in claim 13, wherein:

said obtaining input image data comprises multiplying image data for a first image by a first weighting factor to obtain first intermediate data, multiplying image data for a second image by a second weighting factor to obtain second intermediate data, said first and second intermediate data comprising an intermediate high order number and an intermediate low order number, adding said first and second intermediate low order number, adding said first and second intermediate low order numbers to produce a low order sum and a carry which together correspond to said low order number, and adding said first and second intermediate high order numbers and said carry to produce a high order sum corresponding to said high order number; and

said generating and comparing comprises referencing a look-up table depending on the low order sum from the first adder in order to generate said binary value; and

said outputting comprises incrementing the high order sum depending on the binary value to produce the output image data.

19. An electronic processing apparatus for processing digital image data defining pixels which together represent at least one image, wherein output image data representing each pixel in terms of one of a multiplicity of available characteristic values is produced from input image data representing each pixel in terms of a desired characteristic value which is defined to an accuracy greater than that of the output data such that the desired characteristic value can be defined as a value between available characteristic values, the apparatus comprising:

a source of input image data defining the desired characteristic value of each pixel in terms of a high order number corresponding to one of said multiplicity of available values and a low order number representing the difference between the desired value and said corresponding available value;

a source of data representing a random number;

comparing circuit comparing the random number and the low order number and producing a binary output value depending on the relationship between the random number and the low order number; and

an output circuit outputting as the output image data an available characteristic value selected depending on the value of the high order number and the binary value.