System and method for printing target colors with process colors utilizing parallel feedforward neural networks

• US 8,208,170 B2
• Filed: 10/10/2008
• Issued: 06/26/2012
• Est. Priority Date: 10/10/2008
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Mo, Song-Feng
• Primary Examiner(s): Williams, Kimberly A
• Application Number:
  US12/249,198
• Publication Number:
  US 20100091305A1
• Time in Patent Office: 1,355 Days
• Field of Search: 358/1.9, 358/504, 358/518, 358/520, 358/522, 358/523, 358/524, 382/155, 382/156, 382/157, 382/162, 382/167
• US Class Current: 358/1.9
• CPC Class Codes: H04N 1/6033 using test pattern analysis H04N1/6055 takes precedence

Abstract

A system and method for printing target colors includes a print-engine interface and a neural network component. The print-engine interface is in operative communication with a print engine of a printing system. The neural network component is calibrated to the print engine for printing a target color on a substrate. The neural network is in operative communication with the print-engine interface and communicates a parameter associated with printing the target color on the substrate utilizing the print engine.

Claims

1. A non-transitory computer-readable storage medium for printing target colors, said computer-readable storage medium embodying computer program code, said computer program code comprising computer executable instructions for:

configuring a print-engine interface to be in operative communication with a print engine of a printing system;

calibrating a neural network component to the print engine for printing at least one target color on a substrate, wherein the neural network component includes a plurality of neural networks configured to receive a set of inputs; and

wherein each of the neural networks outputs to the print-engine interface a control parameter of a color separation of the print engine associated with printing the at least one target color on the substrate utilizing the print engine.

2. The non-transitory computer-readable storage medium according to claim 1, wherein the print engine utilizes at least four colorants.

3. The non-transitory computer-readable storage medium according to claim 1, wherein the neural network component includes at least two neural networks arranged in parallel.

4. The non-transitory computer-readable storage medium according to claim 3, wherein one of the at least two networks is a feed-forward neural network.

5. The non-transitory computer-readable storage medium according to claim 1, wherein the neural network component includes at least one three-layer neural network.

6. The non-transitory computer-readable storage medium according to claim 1, wherein the computer-readable storage medium is an installable module installable in the printing system.

7. The non-transitory computer-readable storage medium according to claim 1, wherein the set of inputs includes a point within at least one of a Hunter 1948 L, a, b color space, a CIE 1976 (L*, a*, b*) color space, and a CIE 1931 XYZ color space.

8. The non-transitory computer-readable storage medium according to claim 1, wherein each of the plurality of neural networks outputs a control parameter of a color separation of the print engine, wherein the at least one parameter includes the control parameter.

9. The non-transitory computer-readable storage medium according to claim 8, wherein the control parameter is a Cin value for one of a color separation, a Cyan color separation, a Magenta color separation, a Yellow color separation, and a Black color separation of the print engine.

10. The non-transitory computer-readable storage medium according to claim 1, wherein the only one control parameter is a Cin value for one of a color separation, a Cyan color separation, a Magenta color separation, a Yellow color separation, and a Black color separation of the print engine.

11. The non-transitory computer-readable storage medium according to claim 1, wherein the plurality of neural networks are arranged to form parallel neural networks to control the at least one parameter associated with printing the at least one target color on the substrate utilizing the print engine.

12. The non-transitory computer-readable storage medium according to claim 1, wherein the neural network component comprises:

a calibration component adapted to calibrate a neural network of the neural network component, wherein the calibration component calibrates the neural network to the print engine for printing the at least one target color on the substrate utilizing the print engine.

13. The non-transitory computer-readable storage medium according to claim 12, wherein the calibration component includes a color test pattern analysis component for calibrating the neural network by analyzing at least one color test pattern.

14. The non-transitory computer-readable storage medium according to claim 13, wherein the calibration component is in operative communication with a spectrometer, wherein the spectrometer measures the color test pattern thereby facilitating the color test pattern analysis component to analyze the at least one color test pattern.

15. The non-transitory computer-readable storage medium according to claim 13, wherein the calibration component is in operative communication with a color sensing device, wherein the color sensing device measures the color test pattern thereby facilitating the color test pattern analysis component to analyze the at least one color test pattern, wherein the color sensing device is disposed within the printing system to measure the color test pattern therein.

16. The non-transitory computer-readable storage medium according to claim 12, wherein the calibration component calibrates the neural network to the print engine utilizing at least one of a quick propagation algorithm configured to update a weight of a node of the neural network and a back-propagation algorithm configured for additional adjustment of the weight of the node.

17. A system for printing target colors, comprising:

a print-engine to mark a substrate with a plurality of colorants; and

a plurality of feedfoward neural networks, each of the plurality of feedfoward neural networks being trained to control a single colorant of the plurality of colorants, wherein each of the plurality of feedfoward neural networks is further trained separately to each other and to receive a common input, wherein the plurality of feedfoward neural networks perform parallel integration to control the print-engine to mark the substrate with the plurality of colorants.