Poplar optimization algorithm: A new meta-heuristic optimization technique for numerical optimization and image segmentation

• The POA algorithm mimics the sexual and asexual propagation mechanism of poplar. • The performance of POA is tested on two test suites with different features. • POA is used to find optimal thresholds for image segmentation. • The results indicate that POA is competitive and has its superiority in some cases. A novel algorithm called Poplar Optimization Algorithm (POA) is developed in this paper to solve continuous optimization problems. The algorithm mimics the sexual and asexual propagation mechanism of poplar, where the basic philosophy of how to execute sexual and asexual propagation for individuals is detail designed in the algorithm. Mutation strategy of backtracking search algorithm is adopted in POA to maintain the diversity in a certain degree. The performance of POA algorithm is tested on 25 functions from the CEC2005 test suite and 30 functions from the CEC2017 test suite with different features. The results of POA are compared with some other population-based algorithms in terms of the quality and efficiency. Finally, the proposed algorithm is used to find the optimal threshold for image segmentation. The results indicate that the poplar optimization algorithm can obtain competitive or superior performance.