Evaluation of State of the Art Methods for Segmenting Muscle Cells
Haixing Li, Yongming Yang, Zhenzhou Wang · 2017
Segmentation of muscle cells is important and challenging in microscopy imaging applications. There are many segmentation algorithms available in the literature and they depend on different image features such as pixel intensity value, color and textures. In this paper, three state of the art image segmentation methods: Slope Difference Thresholding and Iterative Erosion based method(SDAIE), SMASH method and CellSegm method were evaluated and compared. To compare the performance of the algorithms and assist the users to understand each method better, different types of muscle cell images and six segmentation evaluation metrics were used. The experimental results showed that the Slope Difference Thresholding and Iterative Erosion based method is more robust than the other two methods.