Dombi entropy based multilevel thresholding of Liver CT scan images using leader slime mould algorithm

One of the efficient methods for segmenting images is multilevel thresholding. In this research paper, a non-parametric based multilevel thresholding (MLT) approach based on Dombi entropy is presented for segmenting Liver CT images. The proposed MLT approach used energy curve instead of the histogram to consider the spatial contextual information for thresholds calculation. As computation time is one of the major challenges in non-parametric based MLT method, this paper used the leader slime mould algorithm (LSMA) to optimize the Dombi entropy based objective function in a reasonable time. LSMA is the modified version of the slime mould algorithm and proved its capability in multilevel thresholding of multispectral images. Real abdominal CT liver images are the benchmark images to assess and test the viability of the suggested multilevel image segmentation technique. The experiment's finding showed that the suggested algorithm is quite promising and can deliver superior segmentation results in comparison to the well-known Tsallis entropy-based thresholding.