White Blood Cells(WBCs) are the major components of the human immune system. Various methodologies are adopted to classify the blood cells to identify infectious diseases. Classification of blood cells is used as a tool to identify infectious diseases. Diagnosis of diseases can easily be obtained by finding out the actual count of their relative frequencies and comparing them with their normal values. The composition of blood is White Blood Cells, Red Blood Cells, Platelets, and Plasma. WBCs, also known as leukocytes, are found in human blood in five different forms. The WBCs play a very important role in defending the body against some infectious diseases. As the WBCs possess distinctive morphological highlights, manual characterization of such cells is a tedious process. Such type of procedure may lead to erroneous calculations as it is for the most part identified with the hematologist’s understanding. These realities accentuate a significant requirement for a quick and computerized strategy for recognizing the distinctive platelets. The multilayer perceptron back-propagation (MLP-BP) neural network is utilised in this study to categorize the most well-known five categories of WBC from blood smear microscopic pictures using the most distinctive attributes. The algorithmic is divided into three stages. Image segmentation is the first step, followed by labeling, which yields the number and location of each WBC, and finally, collecting descriptive information from the segmented cells. The neural network approach is best suited for the analysis of complex data.