C-V level set based cell image segmentation using color filter and morphology

This paper briefly introduces an cell segmentation scheme which is applied to the high resolution cell images. This method is simple, effective and easy to implement. It first uses the C-V level set method to initially separate the cells and other tissues from patches of a cell image, and then an effective color filter is used to determine the cells object and wipe off residual background color. Finally, morphology methods are employed to optimize the segmentation results. The main contribution of our work is that a novel application mode of C-V method is employed to adapt the irregular distribution of cells and complex texture of image. The novelty of our paper is that a new color filter is proposed to eliminate the error mapping from foreground to cells, which is caused by C-V level set method in some cases, and effective morphology process is used to improve the segmentation results. The comparative experiments demonstrate that our method outperforms the original C-V level set method on cells segmentation.