| 摘要 |
A printer operates under a first set of conditions to create a first set of M of samples and under a target set of conditions to create a second set of K samples. Each sample is produced using the same colorant(s). The actual spectral reflectance for each sample is measured. The actual reflectances for corresponding samples in the first and second sets are used to create a cross-validated, partial-least-squares transform that maps a reflectance from a sample produced under the first set of conditions to a reflectance from a sample produced under the target set of conditions. The transform and reflectances produce a set of predicted reflectances, each representing a sample producible under the target set of conditions using one colorant. The predicted reflectances and the colorant(s) are used to generate a table predicting a color value when the printer operates under the target set of conditions.
|
