Color image reproducing system with image signal correction function

Claim 2

A color image signal correction system, comprising: colorimetry means for generating a first colorimetric value signal from a color sample having predetermined color separation values; first image input interface means for receiving a first color separation value signal for the color sample and then for a target color image, and receiving sensor detection of an environmental variation; image output interface means for outputting a second color separation value signal corresponding to the color sample and then to the target color image, and for receiving sensor detection of an environmental variation; color conversion means comprising a neural network for converting the first color separation value signal into a second colorimetric value signal in accordance with a first parameter; color correction means comprising a neural network for converting the first and then the second colorimetric value signals into the second color separation value signal in accordance with a second parameter; and control means for training the neural network of the color conversion means in response to the detection of the environmental variation by the sensor of the first image input interface means so that the first color separation value signal for the color sample is converted into the first colorimetric value signal for the color sample in order to set the first parameter, and training the neural network of the color correction means in response to the detection of the environmental variation by the sensor of the image output interface means so that the first colorimetric value signal for the color sample is converted into the second color separation value signal with the predetermined color separation values in order to set the second parameter.

Claim 3

A system according to claim 1, wherein the colorimetric value signals represent L*a*b* values defined in a CIE L*a*b* space.

Claim 4

A system according to claim 1, wherein the colorimetric value signals represent XYZ values defined in XYZ space.

Claim 5

A system according to claim 1, wherein the colorimetric value signals represent Yxy values defined in a Yxy space.

Claim 6

A system according to claim 1, wherein the colorimetric value signals represent L*u*v* values defined in a L*u*v* space.

Claim 7

A system according to claim 1, wherein the colorimetric value signals represent HVC Munsell values defined in a HVC Munsell space.

Claim 8

A system according to claim 1, wherein the color separation value signals represent CMY values of cyan (C), magenta (M), and yellow (Y).

Claim 9

A system according to claim 1, wherein the color separation value signals represent RGB values of red (R), green (G), and blue (B).

Claim 10

A system according to claim 1, wherein each of the first and second parameters includes a type of the image output means, the layer structure of said neural network, the number of units of each layer in the layer structure, and a weight associated with each unit.

Claim 11

A system according to claim 1, wherein each of said neural networks has at least three layers and each of said layers includes at least three units.

Claim 12

A system according to claim 1, wherein the color sample has a plurality of blocks with colors corresponding to different color separation value signals on the blocks.

Claim 13

An image signal correction system according to claim 1, and further comprising: color image output means connected to the image output interface means for outputting a color image in correspondence with the second color separation value signal and outputting the color sample in correspondence with the predetermined color separation values.

Claim 16

The method for reproducing a target color image according to claim 13 wherein said step of controlling said second neural network includes: a) generating a color sample from said image output device, said color sample corresponding to predetermined color separation value signals; b) colorimetrically measuring said color sample and generating second colorimetric value signals corresponding therewith; c) inputting said second colorimetric value signals into said second neural network device in place of said first set of colorimetric value signals; d) inputting training signals into said second neural network device for adjusting said second parameter until said second set of color separation value signals correspond to said predetermined color separation value signals within a predetermined threshold.

Claim 17

The method for reproducing a target color image according to claim 14 wherein said step of controlling said first neural network includes: substituting said color sample for said target color image; and inputting training signals into said first neural network device for adjusting said first parameter until said first set of colorimetric value signals correspond to said second set of colorimetric value signals within a predetermined threshold.

Claim 18

The method for reproducing a color image according to claim 15 wherein said first parameter comprises a type of the image input device, the layer structure of said neural network, the number of units of each layer in the layer structure, and a weight associated with each unit.

Claim 19

The method for reproducing a target color image according to claim 15, and further comprising storing said adjusted first parameter in a memory storage means.

Claim 20

The method for reproducing a color image according to claim 14 wherein said second parameter includes a type of the image output device, the layer structure of said neural network, the number of units of each layer in the layer structure, and a weight associated with each unit.

Claim 21

The method for reproducing a target color image according to claim 14 and further comprising storing said adjusted second parameter in a memory storage means.

Claim 22

A color image signal correction system, comprising: colorimetry means for generating a first colorimetric value signal from a color sample having predetermined color separation values; first image input interface means for receiving a first color separation value signal for the color sample and then for a target color image, and receiving sensor detection of an environmental variation; second image input interface means for receiving a second colorimetric value signal for the color sample and then for the target color image, and receiving sensor detection of an environmental variation; image output interface means for outputting a second color separation value signal corresponding to the color sample and then to the target color image, and for receiving sensor detection of an environmental variation; color conversion means comprising a neural network for converting the first color separation value signal into a third colorimetric value signal in accordance with a first parameter; color correction means comprising a neural network for converting the first and then the second or third colorimetric value signals into the second color separation value signal in accordance with a second parameter; and control means for selecting one of the second and third colorimetric value signals and, thereby, an operative one of the first and second image input interface means and for training the neural network of the color conversion means in response to the detection of the environmental variation by the sensor of the first image input interface means so that the first color separation value signal for the color sample is converted into the first colorimetric value signal for the color sample in order to set the first parameter, and training the neural network of the color correction means in response to the detection of the environmental variation by the sensor of one of the second image input interface means and the image output interface means so that the first colorimetric value signal for the color sample is converted into the second color separation value signal with the predetermined color separation values in order to set the second parameter.

Claim 23

A color image signal correction system, comprising: colorimetry means for generating a first colorimetric value signal from a color sample having predetermined color separation values; first image input interface means for receiving a first color separation value signal for the color sample and then for a target color image, and receiving sensor detection of an environmental variation; second image input interface means for receiving a second colorimetric value signal for the color sample and then for the target color image, and receiving sensor detection of an environmental variation; image output interface means for outputting a second color separation value signal corresponding to the color sample and then to the target color image, and for receiving sensor detection of an environmental variation; color conversion means comprising a neural network for converting the first color separation value signal into a third colorimetric value signal in accordance with a first parameter; color correction means comprising a neural network for converting the first and then the second or third colorimetric value signals into the second color separation value signal in accordance with a second parameter; and control means for selecting one of the second and third colorimetric value signals and, thereby, an operative one of the first and second image input interface means and for training the neural network of the color conversion means in response to the detection of the environmental variation by the sensor of the first image input interface means so that the first color separation value signal for the color sample is converted into the first colorimetric value signal for the color sample in order to set the first parameter, and training the neural network of the color correction means in response to the detection of the environmental variation by the sensor of one of the second image input interface means and the image output interface means so that the first colorimetric value signal for the color sample is converted into the second color separation value signal with the predetermined color separation values in order to set the second parameter.