返回
Multi-threshold image segmentation algorithm based on Aquila optimization
DOI:10.1007/s00371-023-02993-w.png)
摘要
En 中文
Aquila Optimization (AO) is a recently proposed meta-heuristic algorithm, which has been proved to be more competitive than other meta-heuristic algorithms in function optimization and practical applications. However, when solving more complex optimization problems, AO still has the shortcomings of local optimal stagnation and low solving accuracy. To overcome these shortcomings, an improved Aquila Optimization algorithm (IAO) is proposed in this paper. During the initialization of IAO population, a hybrid chaotic mapping mechanism was introduced to initialize the population, improving both the population diversity and the uniformity of the population distribution. The elite dimensional lens imaging learning strategy is introduced for elite individual to improve the optimization quality of the algorithm as elite individual has more useful information than ordinary individuals. Then the probabilistic jump mechanism of simulated annealing algorithm is used to select the position update mode to balance local development and global search. The experimental results on the CEC2005 test function verify the viability and effectiveness of IAO. IAO is used to the multi-threshold segmentation problem based on symmetric cross entropy to demonstrate its capacity to resolve practical optimization problems. The segmentation performance on different reference images shows that IAO has good segmentation performance in most cases.
Keyword:
Aquila optimization
Hybrid chaotic mapping
Simulated annealing
Lens imaging learning
Symmetric cross-entropy
Muti-threshold segment
期刊
IF:
2.9
论文数:
4.6K
被引数:
6.5K
机构
引用论文
A new feature clustering method based on crocodiles hunting strategy optimization algorithm for classification of MRI images
VISUAL COMPUTER
IF2.9
New color image encryption using hybrid optimization algorithm and Krawtchouk fractional transformations
VISUAL COMPUTER
IF2.9

