White blood cell detection and counting play a crucial role in medical diagnostics, aiding in the diagnosis of various diseases. Traditional manual methods are time-consuming and prone to human error, necessitating the development of automated techniques for accurate and efficient WBC analysis. This study proposes a novel approach leveraging Particle Swarm Optimization to enhance the accuracy and efficiency of WBC detection in microscopic blood images. The method involves converting images to the HSV color space, applying binarization and connected component analysis, and optimizing the results using PSO to eliminate falsely detected regions. Experimental evaluations on a dataset of microscopic blood images demonstrate that the proposed method achieves superior accuracy (61.76%) compared to conventional techniques. The findings suggest that the proposed method is effective in improving WBC detection and counting, making it a valuable tool for automated hematological analysis.
Keywords: Connected Components; HSV Color Space; Microscopic Blood Images; Particle Swarm Optimization; White Blood Cell.