Automated Counting of Platelets and White Blood Cells from Blood Smear Images

摘要

Platelet Detection and Count are one of the major analysis of the pathological test of the blood. Conventional methods of analysis involve observation of blood smear samples under the microscope and manually identifying and counting the numbers. This process is slow and tedious. This work presents a method to automatically detect and count the number of platelets. A sample size of 270 images collected indigenously is used for carrying out the experiments with the proposed methodology, which result in an accuracy of 95.59% for platelets and 100% for WBCs respectively.