The cancer of blood and bone marrow is called Leukemia. It is a result of the uncontrolled reproduction of immature white blood cells. It hampers the ability of the body to fight infection. In Leukemia, the white blood cells (WBC) are generally affected. There are different types of Leukemia namely: Acute myelogenous Leukemia (AML), Acute lymphocytic Leukemia (ALL), Chronic myelogenous Leukemia (CML), Chronic lymphocytic Leukemia (CLL). This paper proposes to automate the Leukemia detection process using machine learning and different techniques of image processing. The dataset consists of images of blood smear which is of both Leukemia and non-Leukemia patients. K-means Clustering, Marker-Controlled Watershed segmentation, and HSV colour-based segmentation are the image segmentation algorithms that have been used. Various features from the segmented lymphocyte images are extracted since the structural components of normal and Leukemic lymphocytes differ remarkably. The SVM classifier, which is a machine learning technique, is used to further classify Leukemia into its different types. This paper aims at identifying Leukemia and determine its types whether it is AML, ALL, CML or CLL which takes the classification process one step further as a majority of the previous works have been restricted to just detection of Leukemia or classifying into few of the main subtypes. The proposed system is successfully implemented using MATLAB.