Here, the value of m is taken as 2, and by experimenting on all possible sizes of cluster, we have achieved best results when the number of clusters is 4. When this algorithm terminates, we obtain a clustered image where each pixel has taken the value of one of the four generated cluster centers. Generally the intensity values of the pixels belonging to the diseased part of a leaf have greater intensity compared to the normal ones. Hence, the rust-spots are assumed to be accommodated in the cluster having the highest intensity value, we give all the pixels in that cluster the maximum available intensity value and we give the minimum to the rest. The binary representation thus obtained will later be used in the Feature Extraction Phase for calculating the number of occurrences of zeros as well as the number of connected components present in the image. The results of Fuzzy c-means Algorithm, applied on some original images are shown in Fig. 3.