Wei Hong Lim

Particle swarm optimization with increasing topology connectivity

In this paper, we propose a new variant of particle swarm optimization (PSO), namely PSO with increasing topology connectivity (PSO-ITC), to solve unconstrained single-objective optimization problems with continuous search space. Specifically, an ITC module is developed to achieve better control of exploration/exploitation searches by linearly increasing the particles topology connectivity with time as well as performing the shuffling mechanism. Furthermore, we introduce a new learning framework that consists of a new velocity update mechanism and a new neighborhood search operator that aims to enhance the algorithms searching performance. The proposed PSO-ITC is extensively evaluated across 20 benchmark functions with various features as well as two engineering design problems. Simulation results reveal that the performance of the PSO-ITC is superior to nine other PSO variants and six cutting-edge metaheuristic search algorithms. The graphical illustration of the proposed particle swarm optimization with increasing topology connectivity (PSO-ITC), consisting of the ITC module and the proposed learning framework.Display Omitted A PSO variant, abbreviated as PSO-ITC, is developed.An ITC module is developed to achieve better balance of global/local searches.A new learning framework is proposed to improve algorithms searching performance.PSO-ITC has prominent searching accuracy and convergence speed in optimization.Results show that PSO-ITC outperforms other PSO variants and MS algorithms.

Teaching and peer-learning particle swarm optimization

The graphical illustration of the proposed teaching and peer-learning PSO (TPLPSO), consisting of the teaching phase, the peer-learning phase, and the stagnation prevention strategy (SPS). A PSO algorithms variant, abbreviated as TPLPSO, is proposed.Teaching and peer-learning framework is proposed to improve PSOs performance.Stagnation prevention strategy is proposed to mitigate the premature convergence.TPLPSO has higher searching accuracy and convergence speed during the optimization.Results show that TPLPSO outperforms other state-of-the-art PSO variants. Most of the recent proposed particle swarm optimization (PSO) algorithms do not offer the alternative learning strategies when the particles fail to improve their fitness during the searching process. Motivated by this fact, we improve the cutting edge teaching-learning-based optimization (TLBO) algorithm and adapt the enhanced framework into the PSO, thereby develop a teaching and peer-learning PSO (TPLPSO) algorithm. To be specific, the TPLPSO adopts two learning phases, namely the teaching and peer-learning phases. The particle firstly enters into the teaching phase and updates its velocity based on its historical best and the global best information. Particle that fails to improve its fitness in the teaching phase then enters into the peer-learning phase, where an exemplar is selected as the guidance particle. Additionally, a stagnation prevention strategy (SPS) is employed to alleviate the premature convergence issue. The proposed TPLPSO is extensively evaluated on 20 benchmark problems with different features, as well as one real-world problem. Experimental results reveal that the TPLPSO exhibits competitive performances when compared with ten other PSO variants and seven state-of-the-art metaheuristic search algorithms.

Novel initialization scheme for Fuzzy C-Means algorithm on color image segmentation

This paper presents a novel initialization scheme to determine the cluster number and obtain the initial cluster centers for Fuzzy C-Means (FCM) algorithm to segment any kind of color images, captured using different consumer electronic products or machine vision systems. The proposed initialization scheme, called Hierarchical Approach (HA), integrates the splitting and merging techniques to obtain the initialization condition for FCM algorithm. Initially, the splitting technique is applied to split the color image into multiple homogeneous regions. Then, the merging technique is employed to obtain the reasonable cluster number for any kind of input images. In addition, the initial cluster centers for FCM algorithm are also obtained. Experimental results demonstrate the proposed HA initialization scheme substantially outperforms other state-of-the-art initialization schemes by obtaining better initialization condition for FCM algorithm.

Particle swarm optimization with adaptive time-varying topology connectivity

A PSO variant, abbreviated as PSO-ATVTC is developed.An ATVTC module is developed to achieve better balance of global/local searches.A new learning framework is proposed to improve algorithms searching performance.PSO-ATVTC has competitive searching accuracy and convergence speed in optimization.Results show that PSO-ATVTC outperforms other PSO variants and MS algorithms. Particle swarm optimization (PSO) has shown its competitive performance for solving benchmark and real-world optimization problems. Nevertheless, it requires better control of exploration/exploitation searches to prevent the premature convergence of swarms. Thus, this paper proposes a new PSO variant called PSO with adaptive time-varying topology connectivity (PSO-ATVTC) that employs an ATVTC module and a new learning framework. The proposed ATVTC module specifically aims to balance the algorithms exploration/exploitation searches by varying the particles topology connectivity with time according to its searching performance. The proposed learning framework consists of a new velocity update mechanism and a new neighborhood search operator to improve the algorithms performance. A comprehensive study was conducted on 24 benchmark functions and one real-world problem. Compared with nine well-established PSO variants and six other cutting-edge metaheuristic search algorithms, the searching performance of PSO-ATVTC was proven to be more prominent in majority of the tested problems.

Two-layer particle swarm optimization with intelligent division of labor

Early studies in particle swarm optimization (PSO) algorithm reveal that the social and cognitive components of swarm, i.e. memory swarm, tend to distribute around the problems optima. Motivated by these findings, we propose a two-layer PSO with intelligent division of labor (TLPSO-IDL) that aims to improve the search capabilities of PSO through the evolution memory swarm. The evolution in TLPSO-IDL is performed sequentially on both the current swarm and the memory swarm. A new learning mechanism is proposed in the former to enhance the swarms exploration capability, whilst an intelligent division of labor (IDL) module is developed in the latter to adaptively divide the swarm into the exploration and exploitation sections. The proposed TLPSO-IDOL algorithm is thoroughly compared with nine well-establish PSO variants on 16 unimodal and multimodal benchmark problems with or without rotation property. Simulation results indicate that the searching capabilities and the convergence speed of TLPSO-IDL are superior to the state-of-art PSO variants.

Color image segmentation using adaptive unsupervised clustering approach

This paper presents the Region Splitting and Merging-Fuzzy C-means Hybrid Algorithm (RFHA), an adaptive unsupervised clustering approach for color image segmentation, which is important in image analysis and in understanding pattern recognition and computer vision field. Histogram thresholding technique is applied in the formation of all possible cells, used to split the image into multiple homogeneous regions. The merging technique is applied to merge perceptually close homogeneous regions and obtain better initialization for the Fuzzy C-means clustering approach. Experimental results have demonstrated that the proposed scheme could obtain promising segmentation results, with 12% average improvement in clustering quality and 63% reduction in classification error compared with other existing segmentation approaches.

Bidirectional teaching and peer-learning particle swarm optimization

Abstract Most of the well-established particle swarm optimization (PSO) variants do not provide alternative learning strategies when particles fail to improve their fitness during the searching process. To solve this issue, we improved the state-of-art teaching–learning-based optimization algorithm and adapted the enhanced framework into the PSO. Thus, we developed a bidirectional teaching and peer-learning PSO (BTPLPSO). Specifically, the BTPLPSO uses two learning phases, namely, the teaching and peer-learning phases. The particles first enter the teaching phase and update their velocity based on their personal and global best information. However, when particles fail to improve their fitness in the teaching phase, they enter the peer-learning phase and learn from the selected exemplar. To establish a two-way learning mechanism between the global best particle and the population, we developed an orthogonal experimental design-based elitist learning strategy to improve the global best particle by fully exploiting the useful information of each particle. The proposed BTPLPSO was thoroughly evaluated on 25 benchmark functions with different characteristics. The simulation results confirmed that BTPLPSO significantly outperforms eight well-established PSO variants and six cutting-edge metaheuristic search algorithms.

Adaptive division of labor particle swarm optimization

A new population division-based particle swarm optimization variant is proposed.Both swarm diversity and fitness are used to adaptively assign the search task of each particles.Two operators are applied on the best solution to further improve the algorithms convergence speed.A stagnation prevention module is also proposed to mitigate the premature convergence issue.The proposed algorithm outperforms its peers in term of searching accuracy and convergence speed. Although evident progress and considerable achievements have been attained in developing a new particle swarm optimization (PSO) algorithm, successfully balancing the exploration and exploitation capabilities of PSO to determine high-quality solutions for complex optimization problems remains a fundamental challenge. In this study, we propose a new PSO variant, namely, adaptive division of labor (ADOL) PSO (ADOLPSO), to overcome the demerits of our previous work. Specifically, an ADOL module is developed in ADOLPSO to adaptively regulate the exploration and exploitation searches of swarm. To achieve this purpose, both criteria of swarm diversity and fitness are considered during the task allocation process of the ADOLPSO current swarm. Two new operators, namely, convex operator and reflectance operator, are adopted to generate new particles from the memory swarm of ADOLPSO to further enhance the searching accuracy and convergence speed of the proposed algorithm. These two operators are activated to evolve the memory swarm only if a fitness improvement is observed in the current swarm of ADOLPSO to prevent excessive computational complexity. The proposed ADOLPSO is applied to solve 18 benchmark functions with various characteristics. Simulation results of ADOLPSO are compared with those of other nine well-established PSO variants. Experimental findings reveal that ADOLPSO significantly outperforms the other PSO variants in terms of searching accuracy, reliability, and convergence speed.

Particle swarm optimization with dual-level task allocation

Abstract Particle swarm optimization (PSO) is a well-known algorithm for global optimization over continuous search spaces. However, this algorithm is limited by the intense conflict between exploration and exploitation search processes. This improper adjustment of exploration and exploitation search processes can introduce an inappropriate level of diversity into the swarm, thereby either decelerating the convergence rate of the algorithm (caused by the excessive diversity) or inducing premature convergence (as a result of insufficient diversity). To address this issue, we propose a new PSO variant, namely, the PSO with dual-level task allocation (PSO–DLTA). Two task allocation modules, that is, the dimension-level task allocation (DTA) and the individual-level task allocation (ITA) modules, are developed in PSO–DLTA to balance the exploration and exploitation search processes. Unlike existing population-based and individual-based task allocation approaches, the DTA module assigns different search strategies to different dimensional components of a particle. Meanwhile, the ITA module serves as an alternative learning phase to enhance the PSO–DLTA particle if it fails to improve in terms of fitness in the DTA module. To demonstrate the effectiveness and efficiency of PSO–DLTA, we compare it with several recently developed optimization algorithms on 25 benchmark and 2 engineering design problems. Experimental results reveal that the proposed PSO–DLTA is more competitive than its contenders in terms of searching accuracy, reliability, and efficiency with respect to most of the tested functions.

Saturation avoidance color correction for digital color images

The qualities of color images captured by digital imaging devices are vulnerable to the scene illumination settings of a given environment. The colors of captured objects may not be accurately reproduced when the illumination settings are uncontrollable or not known a priori. This undesirable property can inevitably degrade the qualities of captured images and lead to difficulties in subsequent image-processing stages. Considering that the task of controlling scene illumination is nontrivial, color correction has emerged as a plausible post-processing procedure to efficiently restore the scene chromatics of a given image. In this study, a new color correction technique called the Saturation Avoidance Color Correction (SACC) algorithm is proposed to remove the undesirable effect of scene illuminants. Unlike most well-established color correction algorithms, the proposed SACC comprises a nonlinear pixel adjustment mechanism to avoid the saturation effect during the color manipulation process. A collection of color images including indoor, outdoor, and underwater images are used to verify the capability of SACC. Extensive experimental studies reveal that the proposed algorithm is preferable to some existing techniques because the former has a high capability to mitigate the color saturation issue and is able to produce corrected images with more pleasant visualization.