An Improved PSO-Based Multilevel Image Segmentation Technique Using Minimum Cross-Entropy Thresholding

Image segmentation is imperative for image processing applications. Thresholding technique is the easiest way of partitioning an image into different regions. Mostly, entropy-based threshold selection methods are used for multilevel thresholding. However, these methods suffer from their dependencies on spatial distribution of gray values. To solve this issue, a novel segmentation error minimization (SEM)-based method for multilevel optimal threshold selection using opposition equilibrium optimizer (OEO) is suggested. In this contribution, a new segmentation score (SS) (objective function) is derived while minimizing the segmentation error function. Our proposal is explicitly free from gray level spatial distribution of an image. Optimal threshold values are achieved by maximizing the SS (fitness value) using OEO. The key to success is the maximization of score among classes, ensuring the sharpening of the shred boundary between classes, leading to an improved threshold selection method. It is empirically demonstrated how the optimal threshold selection is made. Experimental results are presented using standard test images. Standard measures like PSNR, SSIM and FSIM are used for validation The results are compared with state-of-the-art entropy-based technique. Our method performs well both qualitatively and quantitatively. The suggested technique would be useful for biomedical image segmentation.

This paper proposes a novel histogram-based multi-level segmentation scheme of hyper-spectral images. In the proposed scheme an Improved Particle Swarm Optimization (IPSO) algorithm is implemented as a nature-inspired evolutionary algorithm to overcome the drawback of premature convergence and hence getting stuck in local optima problem of PSO. The high-dimension of PSO is decomposed into several one-dimensional problems and premature convergence is removed from each one-dimensional problem. This algorithm is further extended for replacing the worst particles by the fittest particles, determined by their fitness values. Multiple optimal threshold values have been evaluated based on fuzzy-entropy aided with the proposed algorithm. The performance of the IPSO is compared statistically with other global optimization algorithms namely Cuckoo Search (CS), Differential Evolution (DE), FireFly (FF), Genetic Algorithm (GA), and PSO. The produced segmented output of IPSO-fuzzy is then combined with the available ground truth values of image classes to train a Support Vector Machine (SVM) classifier via the composite kernel approach to improving the classification accuracy. This hybrid approach (IPSO-SVM) is then applied to popular hyper-spectral imageries acquired by AVRIS and ROSIS sensors. The final evaluated outcomes of the proposed scheme are also qualitatively compared to show its effectiveness over the other state-of-art global optimizers.

BackgroundGene selection is one of the critical steps in the course of the classification of microarray data. Since particle swarm optimization has no complicated evolutionary operators and fewer parameters need to be adjusted, it has been used increasingly as an effective technique for gene selection. Since particle swarm optimization is apt to converge to local minima which lead to premature convergence, some particle swarm optimization based gene selection methods may select non-optimal genes with high probability. To select predictive genes with low redundancy as well as not filtering out key genes is still a challenge.ResultsTo obtain predictive genes with lower redundancy as well as overcome the deficiencies of traditional particle swarm optimization based gene selection methods, a hybrid gene selection method based on gene scoring strategy and improved particle swarm optimization is proposed in this paper. To select the genes highly related to out samples’ classes, a gene scoring strategy based on randomization and extreme learning machine is proposed to filter much irrelevant genes. With the third-level gene pool established by multiple filter strategy, an improved particle swarm optimization is proposed to perform gene selection. In the improved particle swarm optimization, to decrease the likelihood of the premature of the swarm the Metropolis criterion of simulated annealing algorithm is introduced to update the particles, and the half of the swarm are reinitialized when the swarm is trapped into local minima.ConclusionsCombining the gene scoring strategy with the improved particle swarm optimization, the new method could select functional gene subsets which are significantly sensitive to the samples’ classes. With the few discriminative genes selected by the proposed method, extreme learning machine and support vector machine classifiers achieve much high prediction accuracy on several public microarray data, which in turn verifies the efficiency and effectiveness of the proposed gene selection method.

It is of great significance to study how to use intelligent algorithm to optimize the scheduling of microgrid, so as to improve the operation efficiency of microgrid. In this paper, particle swarm optimization(PSO) algorithm is improved by nonlinearly decreasing inertia weight and nonlinearly dynamic adjusting learning factors. Compared with PSO, the global search ability of the improved particle swarm optimization(IPSO) algorithm is improved. IPSO is used to optimize the scheduling of the microgrid grid connection model. The example results show that nonlinearly decreasing inertia weight and nonlinearly dynamic adjusting learning factors can obviously improve PSO. IPSO has faster convergence and lower comprehensive cost of microgrid scheduling. This paper provides an effective method for the day ahead economic scheduling of microgrid.

For the issues of inferior local search ability and premature convergence in the later evolution stage of the traditional particle swarm optimization, an improved particle swarm optimization is proposed and applied to the parameter estimation. Firstly, in the evolutionary process of particle swarm optimization, the particles which have crossed the border are buffered according to the speed. Then, each particle is performed mutation in different probability according to the evolutionary generations, which can keep the diversity of the particle swarm and avoid the premature convergence effectively. Thirdly, a crossover operation is conducted between the current best particle and the particle which is selected from the particle swarm with a certain probability, which can lead particles gradually approaching to the extreme point and hence, the local search ability of the algorithm will be improved. The advanced particle swarm optimization is applied to the parameter estimation of the kinetic model in the Hg oxidation process and the application result show the effectiveness of the suggested algorithm.

The image segmentation refers to the extraction of region of interest and it plays a vital role in medical image processing. This work proposes multilevel thresholding based on optimization technique for the extraction of region of interest and compression of DICOM images by an improved prediction lossless algorithm for telemedicine applications. The role of compression algorithm is inevitable in data storage and transfer. Compared to the conventional thresholding, multilevel thresholding technique plays an efficient role in image analysis. In this paper, the Particle Swarm Optimization (PSO), Darwinian Particle Swarm Optimization (DPSO), and Fractional Order Darwinian Particle Swarm Optimization (FODPSO) are employed in the estimation of the threshold value. The simulation results reveal that the FODPSO-based multilevel level thresholding generate superior results. The fractional coefficient in FODPSO algorithm makes it effective optimization with fast convergence rate. The classification and blending prediction-based lossless compression algorithm generates efficient results when compared with the JPEG lossy and JPEG lossless approaches. The algorithms are tested for various threshold values and higher value of PSNR indicates the proficiency of the proposed segmentation approach. The performance of the compression algorithms was validated by metrics and was found to be appropriate for data transfer in telemedicine. The algorithms are developed in Matlab2010a and tested on DICOM CT images.

A hybrid differential evolution for optimal multilevel image thresholding

To deal with the problems of premature convergence and tending to jump into the local optimum in the traditional particle swarm optimization, a novel improved particle swarm optimization algorithm was proposed. The self-adaptive inertia weight factor was used to accelerate the converging speed, and chaotic sequences were used to tune the acceleration coefficients for the balance between exploration and exploitation. The performance of the proposed algorithm was tested on four classical multi-objective optimization functions by comparing with the non-dominated sorting genetic algorithm and multi-objective particle swarm optimization algorithm. The results verified the effectiveness of the algorithm, which improved the premature convergence problem with faster convergence rate and strong ability to jump out of local optimum.

Efficient multiprocessor scheduling is essentially the problem of allocating a set of computational jobs to a set of processors to minimize the overall execution time. The main issue is how jobs are partitioned in which total finishing time and waiting time is minimized. Minimization of these two criteria simultaneously, is a multi objective optimization problem. There are many variations of this problem, most of which are NP-hard problem, so we must rely on heuristics to solve the problem instances. Many heuristic-based approaches have been applied to finding schedules that minimize the execution time of computing tasks on parallel processors. Particle swarm optimization (PSO) is currently employed in several optimization and search problems due to its ease and ability to find solutions successfully. A variant of PSO, called as improved particle swarm optimization (ImPSO) has been developed in this paper and is hybridized with the ant colony optimization (ACO) to achieve better solutions. The proposed hybrid algorithm effectively exploits the capabilities of distributed and parallel computing of swarm intelligence approaches. In addition hybrid algorithm using improved particle swarm optimization (ImPSO) with artificial immune system (AIS) is also implemented for the same set of problems to compare with the proposed hybrid algorithm (ImPSO with ACO). It was observed that the proposed hybrid approach (Improved PSO with ACO) gives better results in experiments and reduces finishing and waiting time simultaneously. Key words: Particle swarm optimization (PSO), improved particle swarm optimization (ImPSO), ant colony optimization (ACO), job scheduling, finishing time, waiting time.

To overcome the problem of premature convergence on particle swarm optimization(PSO), an improved particle swarm optimization(IPSO) was proposed, which was called keeping particles active PSO. It was guaranteed to keep the diversity of the particle swarm. When particles lose activity, a special mutation or perturbation was used to activate particles and to make particles explore the search space more efficiently. The IPSO was used to determine optimal substrate feeding rate for fed-batch alcohol fermentation process. The experimental results showed that the IPSO can find better feeding rate quickly than SPSO, and increase alcohol yield more 14 percent than GA.

Optimisation of fixture layout is critical to reduce geometric and form error of the workpiece during the machining process. In this paper the optimal placement of fixture elements (locator and clamp locations) under dynamic conditions is investigated using evolutionary techniques. The application of the newly developed particle swarm optimisation (PSO) algorithm and widely used genetic algorithm (GA) is presented to minimise elastic deformation of the workpiece considering its dynamic response. To improve the performances of GA and PSO, an improved GA (IGA) obtained by basic GA (GA) with sharing and adaptive mutation and an improved PSO (IPSO) obtained by basic PSO (PSO) incorporated into adaptive mutation are developed. ANSYS parametric design language (APDL) of finite element analysis is employed to compute the objective function for a given fixture layout. Three layout optimisation cases derived from the high speed slot milling case are used to test the effectiveness of the GA, IGA, PSO and IPSO based approaches. The comparisons of computational results show that IPSO seems superior to GA, IGA and PSO approaches with respect to the trade-off between global optimisation capability and convergence speed for the presented type problems.

The hybrid energy system design problem needs efficacy tools to reach optimal results in remote areas. The metaheuristic optimization methods are the best choice to address complex problems. This article presents an improved approach based on an energy management strategy for optimal sizing and configuration of standalone photovoltaic scheme components. Improved particle swarm optimization for optimization and configuration of photovoltaic panel and battery system is applied using MATLAB and hourly solar radiation, ambient temperature data, and load demand. The objective of the optimization problem is to define the optimal sizing of the system components to meet the load demand to compose a cost‐effective and reliable scheme. The optimized system results from the improved approach are compared by original particle swarm optimization and simulated annealing algorithms. So, the optimum design of the hybrid scheme is analyzed and compared based on different reliability values. The results show that the improved approach finds the optimal results easily with lower cost, faster convergence, and better reliability indexes in different reliability indexes in comparison with the other approaches. In this regard, for different reliability values (1%, 3%, and 5%), the improved particle swarm optimization shows approximately 22.9% cost saving in comparison with the simulated annealing, and improved particle swarm optimization shows approximately 0.35% cost saving in comparison with the particle swarm optimization. Also, show the reliability of the standalone hybrid photovoltaic system.

In the image segmentation field, it needs several iterations to find optimization thresholds or cluster center to segment images. in this paper, we embedded a scheme based on maximum entropy and uniformity into the particle swarm optimization with sub swarm structure named MEUPSOS to find the optimization threshold values iteratively. Instead of using the conventional PSO, swarm was divided into several sub swarms for the purpose of getting local optimal solutions with a fitness function based on maximal entropy and uniformity. Additionally, just one-swarm particles were used to replace k-swarm (k is the number of threshold values) particles in order to upgrade the computation performance. Then, the local optimal solutions ware used to update global parameter in the global swarm. through iterations updating the velocities and locations of particles, we can calculate the near optimal threshold values on an image based on the fitness functions of maximum entropy and uniformity. Finally, we can find that the proposed method can get more promising results than the other method.

This study aims to improve the real-time accuracy of cooling load forecasting for heating, ventilating and air-conditioning systems (HVAC). This article takes the cooling load in a study room in Qingdao, China, which has been put into use for the first time, as the research object, and establishes a TRNSYS simulation platform to obtain sufficient load data. After using the mean influence value (MIV) and Spearman correlation coefficient to screen the characteristic variables, a hybrid algorithm (CS-CPSO) based on cuckoo search (CS) and particle swarm optimization (PSO) is proposed. Firstly, the iterative extremum is introduced to PSO, secondly, mechanism of levy random flight to generate random new nest in CS is used to initialize PSO particles adaptively, Finally, the optimization algorithm is applied to optimize the back propagation (BP) and support vector regression (SVR) load training models (WBP, WSVR, RBP, RSVR) of the working day (W) and rest day (R), respectively. The maximum grey correlation coefficient is utilized to establish the both models (CS-CPSO-CW, CS-CPSO-CR) of the working day (W) and rest day (R) based on CS-CPSO. In this way, the forecasting results are optimized and then compared with the regression prediction method. The analysis shows that the accuracy of the optimized BP model and SVR model are improved and fully considering the differences, the accuracy of the cooling load prediction is effectively promoted by separately, optimal selection between the prediction values of advanced models (CS-CPSO-WBP, CS-CPSO-WSVR and CS-CPSO-RBP, CS-CPSO-RSVR) gives full play to each algorithm’s advantages and makes up for their shortcomings, and it greatly increases reliability and improves accuracy, which in turn provides the basis for the optimal plan, control, and operation of the HVAC.

The satellite image segmentation is a key area for current research and numerous work has been done for exploration of this area. The nature inspired optimization algorithms are very promising with image segmentation techniques to provide a platform for processing of satellite images. In this paper a literature review of different nature based optimization algorithms such as modified artificial bee colony (MABC) algorithm, ABC algorithm, particle swarm optimization (PSO), Darwinian PSO, genetic algorithm (GA), Wind driven optimization (WDO) and cuckoo search(CS) using different objective functions has been discussed to find the optimized multilevel thresholds. These nature influenced optimization methods and their performances are compared using different objective functions for optimal multilevel thresholding. The comparative study shows that the MABC algorithm and different variants of CS algorithm are very strong and accurate in results generating using image segmentation. Both methods search multilevel thresholds very efficiently and correctly, and in MABC an improved bee’s search solution are used and in CS the cuckoo search solution are used.