In this paper, we describe a new scheme to color image segmentation which is based on supervised pixel classification methods. Using color pixel classification alone does not extract accurately enough color regions, so we suggest to use a strategy based on four steps in different color spaces: simplification, pixel classification, marker extraction and color watershed growing. The strategy is performed on cytological color images. Quantitative measures are used to evaluate the resulting classifications and segmentations with or without a set of reference images.