<![CDATA[White blood cells recognition is developed to help the medical world in diagnosing deseases. In this paper, an artificial neural network algorithm, Fuzzy Learning Vector Quantization is used to recognize the white blood cells. A total of 198 images of white blood cell were used, and were divided into five classes, which are Basofil, Eosinofil, Limfosit, Monosit, and Neutrofil. Experiment was done in 11 types, which are Greyscale, Red, Green, Blue, Hue, Saturation, Value, RGB, HSV, RGB by Voting, and HSV by Voting. Experiment was also done using different value of β and γ, the constant values to increase and decreasing the fuzziness of the reference vectors. In the first experiment the ratio of the training images and testing images were 3:2, and in the second experiment, the ratio were increased to 3:1. Experiment using HSV by Voting has shown a better recognition rate among the others]]>
Copyrights © 2013
