ABSTRACT This study sheds light on a fundamental problem in image segmentation known as multilevel image thresholding. With the rapid growth of artificial intelligence applications that rely on image processing such as medical imaging, remote sensing, and pattern recognition the demand for more effective techniques has become increasingly urgent. Traditional methods suffer from significant limitations, including slow convergence and premature convergence to local optima, particularly when applied to complex or high‐dimensional images. To address these challenges, this study proposes a novel approach based on metaheuristic algorithms, specifically elephant herding optimization (EHO) and symbiotic organism search (SOS). Although these algorithms have shown promising results due to their adaptability and exploratory capabilities, they still face performance bottlenecks resulting from insufficient diversity in the search process. To overcome these limitations, enhanced variants of EHO and SOS are introduced by integrating opposition‐based learning (OBL) and chaos theory to achieve a better balance between exploration and exploitation. These improved algorithms, OCEHO and OCSOS, are applied to the multilevel thresholding problem using Otsu's variance, Kapur's entropy and Masi's entropy as objective functions. The proposed methods are evaluated on 75 standard benchmark images, with segmentation quality evaluated using PSNR, SSIM, and FSIM metrics. Experimental results on 75 standard benchmark images show that the proposed OCEHO algorithm achieves PSNR values up to 37.51 dB, SSIM scores of 0.972, and FSIM values of 0.986, significantly outperforming baseline and hybrid variants. Furthermore, statistical analyzes, including the Wilcoxon rank sum test, confirm the superior stability and convergence speed of OCEHO over its counterparts. These results validate the effectiveness and robustness of the proposed approach for high‐quality image segmentation.

The crayfish optimization algorithm (COA) is a recent bionic optimization technique that mimics the summer sheltering, foraging, and competitive behaviors of crayfish. Although COA has outperformed some classical metaheuristic (MH) algorithms in preliminary studies, it still manifests the shortcomings of falling into local optimal stagnation, slow convergence speed, and exploration–exploitation imbalance in addressing intractable optimization problems. To alleviate these limitations, this study introduces a novel modified crayfish optimization algorithm with multiple search strategies, abbreviated as MCOA. First, specular reflection learning is implemented in the initial iterations to enrich population diversity and broaden the search scope. Then, the location update equation in the exploration procedure of COA is supplanted by the expanded exploration strategy adopted from Aquila optimizer (AO), endowing the proposed algorithm with a more efficient exploration power. Subsequently, the motion characteristics inherent to Lévy flight are embedded into local exploitation to aid the search agent in converging more efficiently toward the global optimum. Finally, a vertical crossover operator is meticulously designed to prevent trapping in local optima and to balance exploration and exploitation more robustly. The proposed MCOA is compared against twelve advanced optimization algorithms and nine similar improved variants on the IEEE CEC2005, CEC2019, and CEC2022 test sets. The experimental results demonstrate the reliable optimization capability of MCOA, which separately achieves the minimum Friedman average ranking values of 1.1304, 1.7000, and 1.3333 on the three test benchmarks. In most test cases, MCOA can outperform other comparison methods regarding solution accuracy, convergence speed, and stability. The practicality of MCOA has been further corroborated through its application to seven engineering design issues and unmanned aerial vehicle (UAV) path planning tasks in complex three‐dimensional environments. Our findings underscore the competitive edge and potential of MCOA for real‐world engineering applications. The source code for MCOA can be accessed at https://doi.org/10.24433/CO.5400731.v1 .

Abstract Nonlinear industrial system modelling entails two critical phases: The first is selecting a method in order to estimate the parameter list values, and the second is selecting a proper model structure with a relatively short parameter list. Developing a comprehensive model for an industrial design process is critical for the model‐based control system. This article presents a model‐based strategy that aims to develop three linear and three nonlinear dynamic models using three well‐known meta‐heuristic optimization algorithms to simulate a challenging plant‐wide process. As a case study, an industrial real winding process (WP) is targeted to accomplish the aim of this study. The algorithms have been optimized to find the best weights of the inputs of the WP with a key issue to effectively describe the behaviour aspects of the process. To test the validity of the developed models, a series of experiments were carried out on each of the developed linear and nonlinear models. Several relevant evaluation metric measures are used to demonstrate the models' performance level. The experimental results for training and test sets of 1250 independent samples for each set based upon the proposed modelling schemes show that the mean square error to correctly model the WP occurred in less than 0.001. A comparison of the developed intelligent linear and nonlinear models with the Auto‐Regressive Integrated Moving Average (ARIMA) and Multiple Linear Regression (MLR) models obtained through the evaluation criteria asserts the effectiveness of the proposed models‐based approaches.

Abstract A population-based optimizer called beluga whale optimization (BWO) depicts behavioral patterns of water aerobics, foraging, and diving whales. BWO runs effectively, nevertheless it retains numerous of deficiencies that has to be strengthened. Premature convergence and a disparity between exploitation and exploration are some of these challenges. Furthermore, the absence of a transfer parameter in the typical BWO when moving from the exploration phase to the exploitation phase has a direct impact on the algorithm’s performance. This work proposes a novel modified BWO (mBWO) optimizer that incorporates an elite evolution strategy, a randomization control factor, and a transition factor between exploitation and exploitation. The elite strategy preserves the top candidates for the subsequent generation so it helps generate effective solutions with meaningful differences between them to prevent settling into local maxima. The elite random mutation improves the search strategy and offers a more crucial exploration ability that prevents stagnation in the local optimum. The mBWO incorporates a controlling factor to direct the algorithm away from the local optima region during the randomization phase of the BWO. Gaussian local mutation (GM) acts on the initial position vector to produce a new location. Because of this, the majority of altered operators are scattered close to the original position, which is comparable to carrying out a local search in a small region. The original method can now depart the local optimal zone because to this modification, which also increases the optimizer’s optimization precision control randomization traverses the search space using random placements, which can lead to stagnation in the local optimal zone. Transition factor (TF) phase are used to make the transitions of the agents from exploration to exploitation gradually concerning the amount of time required. The mBWO undergoes comparison to the original BWO and 10 additional optimizers using 29 CEC2017 functions. Eight engineering problems are addressed by mBWO, involving the design of welded beams, three-bar trusses, tension/compression springs, speed reducers, the best design of industrial refrigeration systems, pressure vessel design challenges, cantilever beam designs, and multi-product batch plants. In both constrained and unconstrained settings, the results of mBWO preformed superior to those of other methods.

Feature selection (FS) is a challenging data mining problem that incorporates a complex search process to find the most informative feature subset. In the brute force methods generating the entire feature space and applying an exhaustive search makes the FS NP-hard problem. Meta-heuristic algorithms are good alternative solutions that provide (near) optimal solutions through a random search process instead of a complete search. In this paper, an FS approach based on the Moth Flame optimization algorithm (MFO) and k-NN classifier are proposed. MFO is a recent meta-heuristic algorithm that has proved its effectiveness in solving different complex problems in a reasonable time. Nevertheless, the performance of MFO highly depends on achieving a balance between exploration and exploitation during the search process. To address this issue, we propose an adaptive method to update the position of a moth toward the best global solution based on the search status. The proposed MFO has been evaluated using sixteen benchmark medical data sets and the results show promising performance of the modified MFO algorithm in terms of the applied evaluation measures.