No abstract

This paper proposes a TSK-type fuzzy controller (TFC) with a two-strategy reinforcement group cooperation based symbiotic evolution (TSR-GCSE) for solving various control problems. The TSR-GCSE proposes the two-strategy reinforcement (TSR) signal designed to improve the performance of the traditional reinforcement signal designed. Moreover, the TSR-GCSE is different from the traditional symbiotic evolution; with each population in the TSR-GCSE method is divided to several groups. Each group represents a set of the chromosomes that belongs to a fuzzy rule and can cooperate with other groups to generation the better chromosomes by using elites-base compensation crossover strategy (ECCS). The illustrative examples show that the proposed method has the better time steps and CPU times than other existing methods.

The artificial blood smear examination is an important disease diagnostics of the patient's condition. However, because the sake of artificial coloring, digital image analysis of blood smear will be suffered by the un-uniform stain which is caused by (1) the procedure of making blood smear, (2) and the different dyeing agent . These situations mentioned above are the reasons that make the quality of image degeneration. In this paper, we present a simple method to ease the influence of un-uniform stain and extract the white blood cell nucleus for further analysis. Therefore, we can simply analyze the blood smear made by hand without using the scanner. Furthermore, we present a feature of lobes to distinguish the nucleus granulocyte (neutrophil, eosinophil, and basophil) from others. The experiment consist of 400 images of each kind of white blood cell to test the accuracy of this system we present.