Multi-threshold image segmentation research based on improved enhanced arithmetic optimization algorithm

Aiming at the shortcomings of arithmetic optimization algorithm (AOA), which has low efficiency and is prone to fall into local optimal solutions, this paper proposes an improved AOA, called IAOA, and applies it to multi-threshold image segmentation processing. Compared with the original algorithm, the improvement points are: Firstly, the AOA introduces inertia weights and cosine factors to enhance the stability of the algorithm, then the chaotic mapping is embedded into the algorithm selection search stage to accelerate the convergence speed, and finally, the strategy of Cauchy mutation and the strategy of inverse learning are introduced to avoid the algorithm from falling into the local optimum. In this paper, 23 benchmark functions in CEC-2005 are used to test the performance of IAOA, and 8 commonly used image segmentation metrics were also selected to test the image segmentation effect of IAOA algorithm. In addition, several heuristic algorithms were also selected for comparison. The results show that compared with other algorithms, IAOA is more robust and has better segmentation performance.