Automatic Segmentation for Ovarian Cancer Immunohistochemical Image Based on YUV Color Space

Segmentation of Immunohistochemical image plays an import part in quantitative assessment of immunohistochemical image. In this paper, an automatic segmentation method was proposed according to characteristics of color immunohistochemical images based on YUV color space. First, it extracts Y, U, V three-component diagram and then we discover that positive cells are distributed in the high gray areas of the V component and distributed in the low gray areas of U component, while the negative cells are located in the high gray areas of the U component and located in the low gray areas of V component. Then, it uses OTSU segmentation method to extract the negative cells region on Y and V components and extract positive cells region on Y and U components. The experimental results show that the method can be a good segmentation of ovarian cancer immunohistochemical images.