Paper Title
A Large-Scale Automated Classification of White Blood Cell Images

Abstract
The counts of several types of white blood cells provide valuable information for diagnosing many diseases. The automation of this task saves time and avoids errors in counting. In this paper, we attempt to classify the white blood cells in peripheral blood based on the shapes and the features of the nucleus. We implement a system and use it to identify and analyze the White Blood Cells (WBCs) automatically. The proposed system can be applied in four steps, namely, segmentation, scanning, feature extraction, and classification of a blood cell. First, we segment the cell images, which involve the categorization of white blood cells into clusters. The second step entails the scanning of each segmented image and preparing the dataset. Extracting the shape and texture from a scanned image is the third step. In the final stage, we apply different machine learning algorithms (SVM, Random Tree, Zero-R) to classify the result based on these criteria. Keywords - Machine learning (ML), Segmentation, Digital image, Image extraction, Histogram