Segmentation of Endothelial Cell Boundaries of Rabbit Aortic Images Using a Machine Learning Approach

Figure 2

Estimated probability density functions of intensities of cell boundary pixels (solid line) and those that lie off the cell boundaries (background, asterisks), using manually labelled, intensity corrected data. The graph illustrates that the pixel intensity distributions of boundary and non-boundary pixels overlap strongly, even over small spatial scales. Adapted from [12].