A Collaborative Dragonfly Algorithm with Novel Communication Strategy and Application for Multi-Thresholding Color Image Segmentation

The Dragonfly Algorithm (DA) is a novel swarm intelligence algorithm with some positive applications in recent years. The algorithm simulates the basic survival ability of dragonflies to evade predators and capture prey in natural environment. The original DA algorithm converges too fast, and it is easy to fall into the local optimum, which causes the search to stagnate and the algorithm effect is not ideal. Based on above, a collaborative evolutionary dragonfly algorithm (CDA) with multi-group strategy is proposed in this paper. It uses multi-group strategy and Cauchy mutation to jointly improve the convergence speed and accuracy of the original algorithm. Image segmentation is an essential aspect of computer graphics and image processing. It has become increasingly important. This paper uses threshold technology based on the CDA algorithm to find the optimal index value under different threshold conditions. The experimental results have demonstrated that the CDA is highly competitive in terms of convergence speed and convergence accuracy DA algorithm, and CDA also performs excellent advantages in graphic segmentation experiments.

Automated summary

This paper introduces the Dragonfly Algorithm and its issues, and proposes an improved collaborative evolutionary Dragonfly Algorithm. Experimental results demonstrate that the algorithm is competitive and performs excellently in terms of graphic segmentation.