Parameter Optimization of Local Fuzzy Patterns for Extracting White Blood Cell Texture Feature

To improve the accuracy, parameter optimization of local fuzzy patterns for extracting white blood cell texture feature is proposed. The proposed method obtains the optimal parameter values of a nucleus and a cytoplasm region of white blood cell image according to search iteratively based on fuzzy contrast measure of nucleus and cytoplasm regions. The proposed method is evaluated using supervised classification of microscopic white blood cell samples images. The results show that the classification accuracy of the local fuzzy pattern features with the optimal parameter values of a nucleus and a cytoplasm region is 5% more accurate than the uniform parameter values and is 5%-20% more accurate than other methods. The proposed method is a part of white blood cell classification; therefore it can be applied for automatic white blood cell differential counting system.