Classification of WBC based on Deep Learning using Microscopic Images
T. Rajalakshmi, C. Senthilkumar · 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS) · 2022
The diagnosis of White Blood cells (WBC) also named as Leukocytes is to classify and detect the total number of WBCs in the human blood cells in order to find the infections, allergies, and diseases. The WBCs are analyzed beneath the microscope, wherein variations in structure and shape reveal the presence of certain diseases. While the examination of WBC images by physicians clinically, a variety of issues may occur due to individual misinterpretations. The Deep Learning method is the best method for accurately and rapidly obtaining white blood cell types. The samples are preprocessed with pixel enhancement, augmentation, normalization, and resizing steps. The Convolutional Neural Network (CNN) method-based results are obtained for WBC classifications. At that point, a Transfer Learning model is used for fine-tuning. As a result, the new model produces classification using the Softmax classifier and finds a higher accuracy equated to other classification approaches.