Fhelo-MTH: Floret Heliotropic Optimized Multilevel Thresholding-based Image Segmentation

The process of segmenting one image into several segments is called image segmentation, which is widely used in several visual understanding applications. Multilevel thresholding-based segmentation enhances the ability to discern subtle changes in the image, making it suitable for detailed image analysis tasks. However, the traditional multilevel segmentation algorithms often pose several challenges including poor segmentation accuracy, stuck in local optimum issues, and minimized convergence speed. Therefore, to tackle the aforementioned challenges, this research proposes a Floret Heliotropic optimized Multilevel Thresholding (FHelO-MTH) algorithm. The integration of the heliotropic movement of sunflowers with the natural pollination process of flowers enhances the FHelO-MTH algorithm to attain optimal threshold values for effective segmentation. Furthermore, the FHelO-MTH algorithm reduces the error rate, improving the proposed algorithm’s overall performance and dependability. When the FHelO-MTH algorithm’s performance is compared to other image segmentation algorithms, the outcomes show a superior Peak signal-to-noise ratio (PSNR) of 45.61 dB.