Stained histological images assist physicians to identify different types of tissues or cells and their architectures. They can be applied on the diagnosis of various diseases and the assessment of treatment effects. Osteoporosis is an aging disease that reduces the density of bones and increases the risk of bone fracture. Literatures indicate that osteoporosis is associated with the ratio of trabecular bone tissues and bone marrow cells, and bones in osteoporosis patients consist of a significantly higher marrow fat content. Interactive segmentation of bone tissue and different types of bone marrow cells in high-resolution histological images, however, is a very tedious and labor-intensive process. The aim of this study is to develop an automatic algorithm to quantify the areas of different tissues such as the trabecular bones and yellow and red marrow cells. This image segmentation method consists of a series of mathematical morphological operation steps based on both the color and morphology features of tissues and was implemented in Matlab. The results obtained from the proposed method have been verified by comparing with those obtained interactively from an experienced histotechnician (Pearson correlation coefficient > 0.94, P < 0.001). The result suggests that the proposed algorithm can effectively assist physicians to quantify stained bone histological images. © 2012 International Society for Advancement of Cytometry.