Multilevel segmentation of Hippocampus images using global steered quantum inspired firefly algorithm

摘要

Microscopic Image segmentation has a crucial role in detecting and diagnosing numerous critical diseases like Alzheimer’s disease, Kidney disease, Cancer, many infectious diseases, etc. Precise segmentation of hippocampus microscopic images is a prerequisite for analyzing and interpreting the brain tissues. A few metaheuristic-based multilevel image segmentation methods are found in the literature for the same. In this work, an enhanced firefly algorithm-based image segmentation method has been proposed to achieve a good quality segmentation. The proposed algorithm utilizes the classical firefly algorithm’s movement operation along with the concept of quantum superposition and quantum update operation. In this algorithm, the movements of quantum fireflies have been modeled based on two strategies: firstly, the less bright fireflies move towards the comparatively brighter ones and secondly, quantum fireflies are updated according to the global optimum by the quantum update operation. This global steered Quantum Inspired Firefly Algorithm (QIFA) has been proposed and used for the multilevel hippocampus image segmentation considering correlation and structural similarity index as objective functions. In order to validate the quality of segmentation, the F-score values with respect to the segmented images have been reported. The proposed algorithm’s performance has been compared with seven other metaheuristic algorithms. The experimental results establish that the proposed algorithm is effective in producing good quality segmentation of hippocampus images.