In the case of digital cameras, device dependent values describe the camera's response to incoming spectrum of light. Transformation from one device space to another has to be defined separately in each case. Device dependent values are not colorimetric and don't necessarily provide a good starting point for transformation between device spaces.We converted the device dependent digital camera RGB values to reflectance spectra, which is used as the device independent color representation. If the spectral power distribution of original and reproduction are identical, a spectral color reproduction is achieved. From spectra, it is possible to calculate response in any color space under arbitrary light sources. We calculated the corresponding results also for direct RGB-CIELAB conversion.In testing phase we modeled the color calibration of a digital camera as a regularized polynomial regression problem. In polynomial regression, strong adaptation of the model to training data can cause problems. Measurement data includes noise that has effect on the complexity of the estimated function, especially when a high order polynomial is used. Effects of overfitting to training data can be dampened by using regularization methods [3]. Two regularization methods, Tikhonov regularization and Truncated Singular Value Decomposition, were tested in order to reduce overfitting.We used Munsell Matte color set (1269 samples) and Macbeth chart (24 samples) in calibration. Analysis of results for different training sets show that the “quality” of the training set is the most important part of the model. As the size of the training set becomes larger, the performance of polynomial model improves. When small training set is used, it must be chosen carefully. With randomly chosen small training sets polynomial model is a very unstable method.