An Optimal Image Segmentation Method based on HGSO for CT Images

Finding the infected regions in medical image modalities is a crucial and challenging task. In this paper, a new image segmentation method is introduced to detect the COVID-19 infection in CT images. In this method, a bi-level-thresholding based image segmentation is proposed using Henry gas solubility optimization. This method used Kapur entropy as a fitness function. Efficiency of the developed segmentation method has been validated on publicly available CT images of COVID-19 patients in terms of PSNR (Pick Signal-to-Noise Ratio), MSE (Mean Square Error), SSIM (Structural Similarity Index Measure) and FSIM (Feature Similarity Index Measure). Moreover, the proposed HGSO-based segmentation method has been compared with SCA, SSA, GWO, CPSOGSA, and MFO-based image segmentation methods to show its efficacy.