Study on Leaf Segmentation Using K-Means and K-Medoid Clustering Algorithm for Identification of Disease

The aim of this paper is to study and compare the analysis of the most commonly used clustering algorithms like K-means and K-medoids based on their basic approach. Clustering is a common method used to segment the images. It is the process of grouping similar objects into different clusters or more precisely, the partitioning of a data set into subsets according to some defined distance measure. It is an unsupervised learning technique, which groups the data in large data sets into different patterns and structures by simple measures. It is widely used in many fields, including machine learning, data mining, pattern recognition, image analysis and bioinformatics. In this research clustering algorithms K-means and K-Medoids were examined and analyzed. From the results, it is found that K-means performs well in situations when the disease is easily divisible from the leaf as it calculate the distance based on the means value.