A Hybrid Approach for Separation of the Strong Adhesion Macrophages

Abstract

The traditional macrophages investigation with microscope is labor-intensive and inefficient. The different degree of adhesion and irregular shapes of macrophages challenge the current image segmentation algorithms, which directly influence the follow-up macrophages analysis. In this paper, we present a hybrid approach for separation of the strong adhesion macrophages, which combines top/bottom hat transformation, gray-scale morphological image processing operations, H-minima transform, watershed algorithm and threshold transform. With strong adhesion macrophages images, we quantitatively demonstrate the performance of our method and provide the comparisons with existing methods. The results show that the hybrid approach can reduce the influence of noise, and obtain precise separation results, because it includes both location and edge information.