Search

Image segmentation methods have received widespread attention in face image recognition, which can divide each pixel in the image into different regions and effectively distinguish the face region from the background for further recognition. Threshold segmentation, a common image segmentation method, suffers from the problem that the computational complexity shows exponential growth with the increase in the segmentation threshold level. Therefore, in order to improve the segmentation quality and obtain the segmentation thresholds more efficiently, a multi-threshold image segmentation framework based on a meta-heuristic optimization technique combined with Kapur's entropy is proposed in this study. A meta-heuristic optimization method based on an improved grey wolf optimizer variant is proposed to optimize the 2D Kapur's entropy of the greyscale and nonlocal mean 2D histograms generated by image computation. In order to verify the advancement of the method, experiments compared with the state-of-the-art method on IEEE CEC2020 and face image segmentation public dataset were conducted in this paper. The proposed method has achieved better results than other methods in various tests at 18 thresholds with an average feature similarity of 0.8792, an average structural similarity of 0.8532, and an average peak signal-to-noise ratio of 24.9 dB. It can be used as an effective tool for face segmentation. [ABSTRACT FROM AUTHOR]

The determination of photovoltaic (PV) parameters is of great importance for the reliability of solar system operation, continuity of the load power consumption, and control management of the energy source. Therefore, this study proposes an advanced backtracking search optimization algorithm (BSA) equipped with teaching and learning‐based optimization (TLBO), named TLBOBSA, to accurately simulate the PV model. During the evaluation of the proposed algorithm, the concept of teaching from TLBO is introduced into the BSA to guide optimal individuals, thus improving the convergence rate of the algorithm. The learning behavior among individuals in the student phase of TLBO facilitates interindividual learning and provides beneficial information for its evolution, which is introduced into the BSA to ensure the diversity of the population. The comprehensive test results of different PV module models in different environmental conditions show that the proposed algorithm is more advantageous for parameter extraction than other existing algorithms. This can be seen in the simulation experiments of two commercial PV models, where the simulated current is consistent with the measured current at each measured voltage. This demonstrates that the proposed TLBOBSA is an accurate and reliable tool for evaluating unknown parameters of PV models. [ABSTRACT FROM AUTHOR]

• An enhanced political optimizer (QNMPO) is proposed to realize the extraction of photovoltaic models parameters. • The performance of the proposed QNMPO algorithm is compared with some well-known competing algorithms. • The stability performance of QNMPO was tested on three commercial PV models at different temperature levels and irradiation levels. • QNMPO exhibits better convergence speed and accuracy. Evaluating the PV system parameters is crucial for the more efficient utilization of sustainable solar energy. Due to PV systems' nonlinear and multimodal nature, efficient evaluation of PV system parameters is still a very challenging task. This paper proposes an improved political optimization algorithm, namely the QNMPO algorithm, which combines quantum rotation gate and Nelder-Mead simplex (NMs). It can be used to evaluate unknown parameters of PV systems and solar cells efficiently. In QNMPO, the quantum rotation gate method can rotate the population of agents to a more favorable position for finding the optimal solution, making the individuals in the middle of the population communicate more frequently and enhancing the population's diversity. In addition, the NMs method can improve the solution quality by better searching the neighborhoods near the optimal solution and then strengthen the advantage of local development of the algorithm so that it can converge to the global optimal solution faster. The developed QNMPO algorithm extracts the unknown parameters for single diode, double diode, triple diode and photovoltaic modules in this study. It is compared with some other representative algorithms. The experimental results show that QNMPO outperforms similar algorithms in convergence speed, solution accuracy, and better solution performance. QNMPO was then used to successfully extract the best parameter values for three real PV models (three commercial PV models (thin-film ST40, monocrystalline SM55, and multi crystalline KC200GT)) at different temperatures and different irradiance levels. The results also show that QNMPO performs consistently and accurately enough under varying temperatures and irradiation. Therefore, the QNMPO algorithm proposed in this paper can be a promising parameter extraction method for solar PV systems with good performance. [ABSTRACT FROM AUTHOR]

• BFGSOLMFO is proposed for global optimization and real-world problems. • OL is used to enhance exploitation and exploration ability of MFO. • BFGS is employed to further excavate the potential global best solution. As a typical emergent swarm intelligence algorithm, Moth-Flame Optimization (MFO) has been created to deal with global optimization problems. Since the introduction, it has been applied to various optimization problems. However, MFO may have the trouble of getting into the local best, and the convergence rate cannot be satisfying when handling the high-dimensional and some multimodal problems. In this work, an enhanced MFO integrated with orthogonal learning (OL) and Broyden-Fletcher-Goldfarb-Shanno (BFGS), which we called BFGSOLMFO, is proposed to alleviate the stagnation shortcomings and accelerate the performance of well-regarded MFO. In the BFGSOLMFO, OL is used to construct a better candidate solution for each moth and then guide the whole population to a reasonable potential area. Meanwhile, in each iteration, after the evolution of population finished and the global optima are obtainable, BFGS is employed to further excavate the potential of the global best moth in the current population. With the aim of evaluating the efficacy of the BFGSOLMFO, first of all, the IEEE CEC2014 benchmark set is utilized to measure the performance in solving function optimizations with high-dimensional and multimodal characteristics. Both sets of the IEEE CEC2011 real-world benchmark problems and the three constrained engineering optimization problems are adopted to estimate the performance of BFGSOLMFO in tackling practical scenarios. In all the experiments, the developed BFGSOLMFO is compared with state-of-the-art advanced algorithms. Experimental results and statistical tests demonstrate that the proposed method outperforms the basic MFO and a comprehensive set of advanced algorithms. [ABSTRACT FROM AUTHOR]

This study proposes a new kernel extreme learning machine (KELM) parameter tuning strategy using a novel swarm intelligence algorithm called grey wolf optimization (GWO). GWO, which simulates the social hierarchy and hunting behavior of grey wolves in nature, is adopted to construct an effective KELM model for bankruptcy prediction. The derived model GWO-KELM is rigorously compared with three competitive KELM methods, which are typical in a comprehensive set of methods including particle swarm optimization-based KELM, genetic algorithm-based KELM, grid-search technique-based KELM, extreme learning machine, improved extreme learning machine, support vector machines and random forest, on two real-life datasets via 10-fold cross validation analysis. Results obtained clearly confirm the superiority of the developed model in terms of classification accuracy (training, validation, test), Type I error, Type II error, area under the receiver operating characteristic curve (AUC) criterion as well as computational time. Therefore, the proposed GWO-KELM prediction model is promising to serve as a powerful early warning tool with excellent performance for bankruptcy prediction. [ABSTRACT FROM AUTHOR]

• This paper proposes an enhanced SCA algorithm variant, called ARSCA. • ARSCA improves the convergence accuracy and avoids premature convergence. • This paper compares ARSCA with classical algorithms, variations of other algorithms. • ARSCA is applied to six engineering design problems and its results are satisfactory. The sine cosine algorithm (SCA) is a well-known meta-heuristic optimization algorithm. SCA has received much attention in various optimization fields due to its simple structure and excellent optimization capabilities. However, the dimension of objective function also increases with the increasing complexity of optimization tasks. This makes the original SCA appear to have insufficient optimization capability and likely to fall into premature convergence. A multi-mechanism acting variant of SCA, called ARSCA, is proposed to address the above deficiencies. ARSCA is an enhanced SCA algorithm based on the adaptive quadratic interpolation mechanism (AQIM) and Rounding mechanism (RM). RM enables a more balanced state between exploration and exploitation of the ARSCA. AQIM enhances local exploitation capabilities. To verify the performance of ARSCA, we compared ARSCA with some advanced traditional optimization algorithms and variants of algorithms for 30 consecutive benchmark functions of IEEE CEC2014. In addition, ARSCA was applied to 6 constrained engineering optimization problems. These six algorithms include the tension–compression spring design problem, the welded beam design problem, the pressure vessel design problem, the I-beam design problem, the speed reducer design problem, and the three-bar design problem. Experimental results show that ARSCA outperforms its competitors in both the solution quality and the ability to jump out of the local optimum. The relevant codes for the paper are publicly available at https://github.com/YangXiao9799/paper_ARSCA. [ABSTRACT FROM AUTHOR]

• This paper proposes an enhanced SCA algorithm variant, called ARSCA. • ARSCA improves the convergence accuracy and avoids premature convergence. • This paper compares ARSCA with classical algorithms, variations of other algorithms. • ARSCA is applied to six engineering design problems and its results are satisfactory. The sine cosine algorithm (SCA) is a well-known meta-heuristic optimization algorithm. SCA has received much attention in various optimization fields due to its simple structure and excellent optimization capabilities. However, the dimension of objective function also increases with the increasing complexity of optimization tasks. This makes the original SCA appear to have insufficient optimization capability and likely to fall into premature convergence. A multi-mechanism acting variant of SCA, called ARSCA, is proposed to address the above deficiencies. ARSCA is an enhanced SCA algorithm based on the adaptive quadratic interpolation mechanism (AQIM) and Rounding mechanism (RM). RM enables a more balanced state between exploration and exploitation of the ARSCA. AQIM enhances local exploitation capabilities. To verify the performance of ARSCA, we compared ARSCA with some advanced traditional optimization algorithms and variants of algorithms for 30 consecutive benchmark functions of IEEE CEC2014. In addition, ARSCA was applied to 6 constrained engineering optimization problems. These six algorithms include the tension–compression spring design problem, the welded beam design problem, the pressure vessel design problem, the I-beam design problem, the speed reducer design problem, and the three-bar design problem. Experimental results show that ARSCA outperforms its competitors in both the solution quality and the ability to jump out of the local optimum. The relevant codes for the paper are publicly available at https://github.com/YangXiao9799/paper_ARSCA. [ABSTRACT FROM AUTHOR]