IFODPSO-based multi-level image segmentation scheme aided with Masi entropy

This article presents an improved version of Fractional Order Darwinian PSO (IFODPSO) for segmenting 3D histogram-based color images at multiple levels of Berkley Segmentation Dataset (BSDS500). The success of convergence and accuracy rate of FODPSO algorithm depends on the value of fractional coefficient. This concept may provide drawback to the algorithm specially for multilevel problems of large dataset. So, to overcome the full dependency on fractional coefficient, delta potential model of quantum mechanics has been incorporated with FODPSO for updating the particle's present as well as global position by destroying the worst particles (solutions), formulated using the introduction of the context parameter. Multi-level Massi Entropy (MME), of current interest, has been chosen here as the objective function for finding the threshold values in combination with IFODPSO. Further, the small segmented regions have been removed or merged into bigger regions for showing the better discrimination between different segmented objects. The effectiveness of the proposed MME-IFODPSO algorithm has been extensively investigated in terms of statistically and qualitatively in terms of the fidelity parameters with the state-of-art approaches and it has been found that the proposed method has improved at least 2-5% to the conventional methods in terms of accuracy.

Automated summary

This article introduces an improved Fractional Order Darwinian PSO (IFODPSO) algorithm for color image segmentation, which reduces the dependence on fractional coefficient by incorporating the delta potential model of quantum mechanics. The algorithm achieves better segmentation accuracy by utilizing Multi-level Massi Entropy as the objective function.