Giant Trevally Optimizer (GTO) is an intelligent optimization algorithm constructed based on the unique hunting behavior of seabirds by giant trevally. A modified Giant Trevally Optimizer (MGTO) was proposed to address the issues of uneven initial population distribution and premature local optima in the MGTO, and it was applied in the fields of medical and plant image segmentation. In order to improve the diversity of the initial population, logistic chaotic mapping was introduced to improve the population initialization mode of MGTO, and a reverse learning strategy was adopted to improve its search performance and avoid the MGTO falling into local optima. Finally, with Kapur as the objective function, a comparative analysis was conducted with GTO, Modified Whale Optimization Algorithm (MWOA), and Satin Bowerbird Optimizer Algorithm (SBO) based on a large amount of visualization and quantitative data analysis. The results indicate that the MGTO has achieved better segmentation results in the field of image segmentation.