RBC Image Segmentation Based on Shape Reconstruction and Multi-scale Surface Fitting

The shape deformability of Red Blood Cell plays an important role in blood filterability. This paper proposed a combined approach for complex surface segmentation of Red Blood Cell. Firstly the individual cell was extracted using contour guiding method. Each pixel's shading variation information was then used for 3D shape reconstruction. The recovered depth map of each image contributed to segment the shape through multi-scale surface fitting. The experimental result shown good results and obtained high classification accuracy.