Segmentation of Plant Images plays an important role in modern agriculture where it can provide accurate analysis of a plant’s growth and possible anomalies. In this paper, rough set based partitional clustering technique called Rough K-Means has been utilized in CIELab color space for the proper leaf segmentation of rosette plants. The efficacy of the proposed technique have been analysed by comparing it with the results of traditional K-Means and Fuzzy C-Means clustering algorithms. The visual and numerical results reveal that the RKM in CIELab provides the nearest result to the ideal ground truth, hence the most efficient one.