# US 5,353,127 A · Method for quantization gray level pixel data with extended distribution set > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 5,353,127 A - Document ID: US5353127A - Status: Granted - Kind: A - Application number: 08/167,758 - Filed: 1993-12-15 - Granted: 1994-10-04 - Earliest claimed: 1993-12-15 - Assignees: Xerox Corporation - Inventors: Jeng-Nan Shiau; Zhigang Fan - Web page: https://patentgrep.com/patent/US5353127A ## CPC - H04N 1/4052: {by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions} - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A method and arrangement for quantization of the nth pixel in scan line l (i.e., pixel n,l), with error diffusion to a matrix including pixels at pixel locations corresponding to pixels {(n+1,l), (n-2,l+1), (n-1,l+1), (n,l+1)}. Distribution weighting may be in accordance with the scheme {(n+1,l)=0.5, (n-2,l+1)=0.125 (n-1,l+1)=0.125 (n,l+1)= 0.25 }. In another embodiment, distribution weighting may be in accordance with the scheme {(n+1,l)=0.5, (n-3,l+1)=0.0625, (n-2,l+1)=0.0625, (n-1,l+1)=0.125, (n,l+1)=0.25, }. ## Claims ### Claim 1 A method of quantizing pixel values, in which an image is formed by a plurality of pixels, each pixel representing an optical density of the image at a location within the image, and having an original optical density value associated therewith selected from one of a set of original optical density values, having number of members larger than a desired output set of optical density values, the steps comprising: adding to each pixel an error resulting from quantization of previous pixels, if any, to derive a modified optical density value; comparing the modified optical density value of an n.sup.th pixel in a scan line l (n,l) with a threshold to select an output values representing the pixel; determining an error term that is a difference between the output value and the modified optical density value; distributing a weighted portion of the error term, to a preselected plurality of pixels, including pixels at positions (n+1,l), (n-2,l+1), (n-1,l+1), (n,l+1). ### Claim 2 (depends on 1) The method as described in claim 1, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-2,l+1)=0.125; (n-1,l+1)=0.125; and (n,l+1)=0.25. ### Claim 3 (depends on 1) The method as described in claim 1, wherein the preselected plurality of pixels additionally includes pixel (n-3,l+1). ### Claim 4 (depends on 3) The method as described in claim 3, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-3,l+1)=0.0625; (n-2,l+1)=0.0625; (n-1,l+1)=0.125; (n,l+1)=0.25. ### Claim 5 A method of preparing an electronic document for printing at a printer adapted to printing n level spots for each pixel, the steps comprising: receiving a document defined in terms of digital electronic input signals, each signal having a magnitude selected from one of m levels (where m is greater than n) and corresponding to a pixel; adding to each input signal an error signal resulting from quantization of previous signals, if any, to derive a modified input signal; comparing the modified input signal of an n.sup.th pixel in scan line l (n,l) with a threshold to select as an output response from the printer, one of the n level spots; determining an error term that is a difference between the output response and the modified input for the n.sup.th pixel in scan line l (n,l); storing the error term; distributing a weighted portion of the error term, to a preselected plurality of input signals, including signals representing density in the document at positions (n+1,l), (n-2,l+1), (n-1,l+1), (n,l+1). ### Claim 6 (depends on 5) The method as described in claim 5, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-2,l+1)=0.125; (n-1,l+1)=0.125; and (n,l+1)=0.25. ### Claim 7 (depends on 5) The method as described in claim 5, wherein the preselected plurality of pixels additionally includes pixel (n-3,l+1). ### Claim 8 (depends on 7) The method as described in claim 7, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-3,l+1)=0.0625; (n-2,l+1)=0.0625; (n-1,l+1)=0.125; (n,l+1)=0.25. ### Claim 9 A method of quantizing pixel values, in which an image is formed by a plurality of pixels, each pixel representing an optical density of the image at a location therewithin, and having an original optical density value associated therewith selected from one of a set of original optical density values, having number of members n larger than a desired output set of optical density values, the steps comprising: adding to an input density value of each pixel an error signal resulting from quantization of previous pixels, if any, to derive a modified density value; comparing the modified density value of an n.sup.th pixel in a scan line l (n,l) with a threshold to select an output value representing the pixel from a set of m possible density value; determining an error term that is a difference between the output value and the modified density value; distributing a weighted portion of the error term, to a preselected plurality of pixels including a pixels representing density at (n+1,l), and pixels in the next scan line at pixel positions less than n. ### Claim 10 (depends on 9) The method as described in claim 9, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-2,l+1)=0.125; (n-1,l+1)=0.125; and (n,l+1)=0.25. ### Claim 11 (depends on 9) The method as described in claim 9, wherein the preselected plurality of pixels additionally includes pixel (n-3,l+1). ### Claim 12 (depends on 11) The method as described in claim 11, wherein the weighting of each weighted portion corresponds to: (n+1,l)=0.5; (n-3,l+1)=0.0625; (n-2,l+1)=0.0625;(n-1,l+1)=0,125; (n,l+1)=0.25. ### Claim 13 Apparatus for preparing an electronic document for printing at a printer adapted to printing n level spots for each pixel, the steps comprising: a document input receiving document images defined in terms of digital electronic input signals, each signal having a magnitude selected from one of m levels (where m is greater than n) and corresponding to a pixel and describing density of the document at position n in a scan line l (n,l); a summing circuit, adding to each input signal an error signal resulting from quantization of previous signals, if any, to derive a modified input signal; a comparator, operatively connected to said document input to receive the modified input signals and to a source of threshold signals, and producing an output response selected from one of the n level spots upon comparing each modified input signal with a threshold signal; a difference circuit receiving as inputs the output response and the modified input signals and deriving an digital electronic error term signal representing the difference in density between the output response and the modified input signal; a first error term buffer and a corresponding shift register operatively connected to said difference circuit, receiving an error term from quantization of signal (n-1,l) therefrom and shifting the error term for addition of a weighted portion thereof at said summing circuit to input signal n in scan line l (n,l); a line buffer operatively connected to the first error term buffer, storing a line of error term signals corresponding to signals in scan line l-1; at least second, third and fourth error term buffers, each having a corresponding shift register, and operatively connected to said line buffer, to receive error terms from the quantization of signals at positions (n,l-1), (n+1,l-1), (n+2,l-1) each shift register shifting the error term for addition of a weighted portion thereof at said summing circuit to input signal n in scan line l (n,l). ### Claim 14 (depends on 13) The apparatus as described in claim 13, wherein the weighted portions of the error term from positions (n,l-1), (n+1,l-1), (n+2,l-1) corresponds to: (n-1,l)=0.5; (n+2,l-1)=0.125; (n+1,l-1)=0.125; and (n,l-1)=0.25. ### Claim 15 (depends on 13) The apparatus as described in claim 13, including a fifth error term buffer and corresponding shift register, and operatively connected to said line buffer, to receive error terms from the quantization of a signal at position (n+3, l-1), each shift register shifting the error term for addition of a weighted portion thereof at said summing circuit to input signal n in scan line l (n,l). ### Claim 16 (depends on 15) The apparatus as described in claim 15, wherein the weighting of each weighted portion corresponds to: (n-1,l)=0.5; (n+3,l-1)=0.0625; (n+2,l-1)=0.0625; (n+1,l-1)=0.125; (n,l-1)=0.25.