Spectral reflection prediction models, although effective, are impractical for certain industrial applications such as self-calibrating devices and online monitoring because of the requirements imposed by their calibration. The idea emerged to make the calibration more flexible. Instead of requiring specific color-constant calibration patches, the calibration would rely on the information contained in regular prints, e.g. on information found in printed color images. Using the CMYK Ink Spreading enhanced Yule-Nielsen modified Spectral Neugebauer model (IS-YNSN), the objective of this dissertation is to recover the Neugebauer primaries and ink spreading curves from image tiles extracted from printed color images. The IS-YNSN is first reviewed in the context of CMYK prints. Two sources of ambiguity are identified and removed, yielding a more robust model better suited for a flexible calibration. We then propose a gradient-descent method to acquire the ink spreading curves from image tiles by relying on constraints based on a metric evaluating the relevance of each ink spreading curve to the set of image calibration tiles. We optimize the algorithm which automatically selects the image tiles to be extracted and show that 5 to 10 well-chosen image tiles are sufficient to accurately acquire all the ink spreading curves. The flexible calibration is then extended to recover the Neugebauer primaries from printed color images. Again, a simple gradient-descent algorithm is not sufficient. Thanks to a set of constraints based on Principal Component Analysis (PCA) and the relationships between composed Neugebauer primaries and the ink transmittances, good approximations of the Neugebauer primaries are achieved. These approximations are then optimized, yielding an accurately calibrated IS-YNSN model comparable to one obtained by classical calibrations. A detailed analysis of these calibrations shows that 25 well-chosen CMYK image calibration tiles are sufficient to accurately recover both the Neugebauer primaries and the ink spreading curves.