Qualitative Abnormalities of Peripheral Blood Smear Images Using Deep Learning Techniques

In recent years, Peripheral blood smear is a generic analysis to assess the person’s health status. Manual testing of images are difficult, time-consuming and subject human intervention visual error. This method encouraged for researchers present algorithms techniques perform peripheral with help computer-assisted decision-making techniques. Existing CAD based methods lacks in attaining accurate detection abnormalities images. order mitigate this issue Deep Convolution Neural Network (DCNN) automatic classification technique introduced eight groups cells such as basophil, eosinophil, lymphocyte, monocyte, neutrophil, erythroblast, platelet, myocyte, promyocyte metamyocyte. The proposed DCNN model employs transfer learning approach additionally it carries three stages pre-processing, feature extraction classification. Initially pre-processing steps incorporated eliminate noisy contents image by using Histogram Equalization (HE). It enclosed improve an contrast. distinguish dissimilar class segmentation carried out Fuzzy C-Means (FCM) whereas its centroid point optimality Slap Swarm optimization strategy. Moreover some specific set Gray Level Co-occurrence Matrix (GLCM) features segmented extracted augment performance algorithm. Finally recorded classifier has capability extract their own features. Based on diverse classes classified distinguished from qualitative found image.