scholarly journals Parallel Compact Differential Evolution for Optimization Applied to Image Segmentation

2020 ◽  
Vol 10 (6) ◽  
pp. 2195
Author(s):  
Xiao Sui ◽  
Shu-Chuan Chu ◽  
Jeng-Shyang Pan ◽  
Hao Luo
Keyword(s):  
Image Segmentation ◽  
Differential Evolution ◽  
Communication Strategies ◽  
Optimal Solution ◽  
Mean Value ◽  
Computing Power ◽  
Elite Strategy ◽  
Local Optimal Solution ◽  
The Individual

A parallel compact Differential Evolution (pcDE) algorithm is proposed in this paper. The population is separated into multiple groups and the individual is run by using the method of compact Differential Evolution. The communication is implemented after predefined iterations. Two communication strategies are proposed in this paper. The first one is to replace the local optimal solution by global optimal solution in all groups, which is called optimal elite strategy (oe); the second one is to replace the local optimal solution by mean value of the local optimal solution in all groups, which is called mean elite strategy (me). Considering that the pcDE algorithm does not need to store a large number of solutions, the algorithm can adapt to the environment with weak computing power. In order to prove the feasibility of pcDE, several groups of comparative experiments are carried out. Simulation results based on the 25 test functions demonstrate the efficacy of the proposed two communication strategies for the pcDE. Finally, the proposed pcDE is applied to image segmentation and experimental results also demonstrate the superior quality of the pcDE compared with some existing methods.

Research and Improvement on K-Means Clustering Algorithm

2013 ◽  
Vol 756-759 ◽  
pp. 3231-3235
Author(s):  
Xue Mei Wang ◽  
Jin Bo Wang
Keyword(s):  
Differential Evolution ◽  
Clustering Algorithm ◽  
Differential Evolution Algorithm ◽  
Optimal Solution ◽  
Clustering Quality ◽  
Local Optimal Solution ◽  
Evolution Algorithm ◽  
Clustering Center ◽  
Searching Ability ◽  
Improved Differential Evolution Algorithm

According to the defects of classical k-means clustering algorithm such as sensitive to the initial clustering center selection, the poor global search ability, falling into the local optimal solution. A differential evolution algorithm which was a kind of a heuristic global optimization algorithm based on population was introduced in this article, then put forward an improved differential evolution algorithm combined with k-means clustering algorithm at the same time. The experiments showed that the method has solved initial centers optimization problem of k-means clustering algorithm well, had a better searching ability,and more effectively improved clustering quality and convergence speed.

A Image Segmentation Method of Improved Ant Colony Algorithm for the Manipulator Self-Recognition Target

2011 ◽  
Vol 135-136 ◽  
pp. 50-55
Author(s):  
Yuan Bin Hou ◽  
Yang Meng ◽  
Jin Bo Mao
Keyword(s):  
Image Segmentation ◽  
Ant Colony Algorithm ◽  
Optimal Solution ◽  
Immune Algorithm ◽  
Ant Colony ◽  
Segmentation Result ◽  
Segmentation Method ◽  
Self Recognition ◽  
Local Optimal Solution ◽  
Improved Ant Colony Algorithm

According to the requirements of efficient image segmentation for the manipulator self-recognition target, a method of image segmentation based on improved ant colony algorithm is proposed in the paper. In order to avoid segmentation errors by local optimal solution and the stagnation of convergence, ant colony algorithm combined with immune algorithm are taken to traversing the whole image, which uses pheromone as standard. Further, immunization selection through vaccination optimizes the heuristic information, then it improves the efficiency of ergodic process, and shortens the time of segmentation effectively. Simulation and experimental of image segmentation result shows that this algorithm can get better effect than generic ant colony algorithm, at the same condition, segmentation time is shortened by 6.8%.

scholarly journals An Improved Grey Wolf Optimizer Based on Differential Evolution and OTSU Algorithm

2020 ◽  
Vol 10 (18) ◽  
pp. 6343
Author(s):  
Yuanyuan Liu ◽  
Jiahui Sun ◽  
Haiye Yu ◽  
Yueyong Wang ◽  
Xiaokang Zhou
Keyword(s):  
Differential Evolution ◽  
Image Recognition ◽  
Optimal Solution ◽  
Tsallis Entropy ◽  
Grey Wolf Optimizer ◽  
Initial Population ◽  
Grey Wolf ◽  
Particle Swarm Optimizer ◽  
De Algorithm ◽  
Local Optimal Solution

Aimed at solving the problems of poor stability and easily falling into the local optimal solution in the grey wolf optimizer (GWO) algorithm, an improved GWO algorithm based on the differential evolution (DE) algorithm and the OTSU algorithm is proposed (DE-OTSU-GWO). The multithreshold OTSU, Tsallis entropy, and DE algorithm are combined with the GWO algorithm. The multithreshold OTSU algorithm is used to calculate the fitness of the initial population. The population is updated using the GWO algorithm and the DE algorithm through the Tsallis entropy algorithm for crossover steps. Multithreshold OTSU calculates the fitness in the initial population and makes the initial stage basically stable. Tsallis entropy calculates the fitness quickly. The DE algorithm can solve the local optimal solution of GWO. The performance of the DE-OTSU-GWO algorithm was tested using a CEC2005 benchmark function (23 test functions). Compared with existing particle swarm optimizer (PSO) and GWO algorithms, the experimental results showed that the DE-OTSU-GWO algorithm is more stable and accurate in solving functions. In addition, compared with other algorithms, a convergence behavior analysis proved the high quality of the DE-OTSU-GWO algorithm. In the results of classical agricultural image recognition problems, compared with GWO, PSO, DE-GWO, and 2D-OTSU-FA, the DE-OTSU-GWO algorithm had accuracy in straw image recognition and is applicable to practical problems. The OTSU algorithm improves the accuracy of the overall algorithm while increasing the running time. After adding the DE algorithm, the time complexity will increase, but the solution time can be shortened. Compared with GWO, DE-GWO, PSO, and 2D-OTSU-FA, the DE-OTSU-GWO algorithm has better results in segmentation assessment.

Research on the Optimal Design of Soccer Robot Based on the Mechanical Analysis

2013 ◽  
Vol 380-384 ◽  
pp. 3854-3857
Author(s):  
Jian Wen Han ◽  
Lei Hong
Keyword(s):  
Differential Evolution ◽  
Clustering Algorithm ◽  
Differential Evolution Algorithm ◽  
Optimal Solution ◽  
Mechanical Analysis ◽  
Clustering Quality ◽  
Local Optimal Solution ◽  
Evolution Algorithm ◽  
Searching Ability ◽  
Improved Differential Evolution Algorithm

According to the defects of classical k-means clustering algorithm such as sensitive to the initial clustering center selection, the poor global search ability, falling into the local optimal solution. A differential evolution algorithm which was a kind of a heuristic global optimization algorithm based on population was introduced in this article, then put forward an improved differential evolution algorithm combined with k-means clustering algorithm at the same time. The experiments showed that the method has solved initial centers optimization problem of k-means clustering algorithm well, had a better searching ability,and more effectively improved clustering quality and convergence speed

The Application of Quantum-Behaved Particle Swarm Algorithm in Intelligent Test Paper Generation

2013 ◽  
Vol 694-697 ◽  
pp. 2378-2382 ◽  
Author(s):  
Xin Ran Li
Keyword(s):  
Particle Swarm ◽  
Optimal Solution ◽  
Particle Aggregation ◽  
Particle Swarm Algorithm ◽  
Test Paper ◽  
Inertia Weight ◽  
Local Optimal Solution ◽  
Swarm Algorithm ◽  
Low Efficiency

Aiming at solving the low efficiency and low quality of the existing test paper generation algorithm, this paper proposes an improved particle swarm algorithm, a new algorithm for intelligent test paper generation. Firstly, the paper conducts mathematically modeling based on item response theory. Secondly, in the new algorithm, the inertia weight is expressed as functions of particle evolution velocity and particle aggregation by defining particle evolution velocity and particle aggregation so that the inertia weight has adaptability. At the same time, slowly varying function is introduced to the traditional location updating formula so that the local optimal solution can be effectively overcome. Finally, simulation results show that compared with the quantum-behaved particle swarm algorithm, the proposed algorithm has better performance in success rate and composing efficiency.

Research on Logistics Distribution Center Location Based on Improved Immune Algorithm

2013 ◽  
Vol 756-759 ◽  
pp. 1366-1370
Author(s):  
Hong Li ◽  
Xiao Yu Liu ◽  
Jin Ping Zhang ◽  
Ya Long Yu
Keyword(s):  
Optimal Solution ◽  
Immune Algorithm ◽  
Distribution Center ◽  
Logistics System ◽  
Logistics Center ◽  
Center Location ◽  
Elite Strategy ◽  
Local Optimal Solution ◽  
Distribution Center Location

Logistics center is a mordern logistic facility. The distribution center location determines the operational efficiency of logistics system. The optimum location of distribution center is important to transportation. In order to improved the algorithm's efficiency, elite strategy was introduced based on the standard immune algorithm. The improved algorithm avoid trapping into local optimal solution and solving the problem more slowly. The role of the elite strategy is to make the optimal solution attractively in the next cycle. This method sovles problem both quantitatively and qualitatively, which makes final solution better in accordance with practical demands.

scholarly journals Multipopulation Genetic Algorithm Based on GPU for Solving TSP Problem

2020 ◽  
Vol 2020 ◽  
pp. 1-8
Author(s):  
Boqun Wang ◽  
Hailong Zhang ◽  
Jun Nie ◽  
Jie Wang ◽  
Xinchen Ye ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Optimal Solution ◽  
Coarse Grained ◽  
Mutation Rates ◽  
Adaptive Mutation ◽  
Original Population ◽  
Elite Strategy ◽  
Mutation Strategy ◽  
Local Optimal Solution ◽  
Crossover Method

A GPU-based Multigroup Genetic Algorithm was proposed, which parallelized the traditional genetic algorithm with a coarse-grained architecture island model. The original population is divided into several subpopulations to simulate different living environments, thus increasing species richness. For each subpopulation, different mutation rates were adopted, and the crossover results were optimized by combining the crossover method based on distance. The adaptive mutation strategy based on the number of generations was adopted to prevent the algorithm from falling into the local optimal solution. An elite strategy was adopted for outstanding individuals to retain their superior genes. The algorithm was implemented with CUDA/C, combined with the powerful parallel computing capabilities of GPUs, which greatly improved the computing efficiency. It provided a new solution to the TSP problem.

Stratified Sampling Differential Evolution Algorithm for Global Optimization Problem

2012 ◽  
Vol 452-453 ◽  
pp. 1491-1495
Author(s):  
Shu Hua Wen ◽  
Qing Bo Lu ◽  
Xue Liang Zhang
Keyword(s):  
Differential Evolution ◽  
Optimization Problem ◽  
Sampling Method ◽  
Differential Evolution Algorithm ◽  
Optimal Solution ◽  
Stratified Sampling ◽  
Local Optima ◽  
De Algorithm ◽  
Local Optimal Solution ◽  
Evolution Algorithm

Differential Evolution (DE) is one kind of evolution algorithm, which based on difference of individuals. DE has exhibited good performance on optimization problem. However, when a local optimal solution is reached with classical Differential Evolution, all individuals in the population gather around it, and escaping from these local optima becomes difficult. To avoid premature convergence of DE, we present in this paper a novel variant of DE algorithm, called SSDE, which uses the stratified sampling method to replace the random sampling method. The proposed SSDE algorithm is compared with some variant DE. The numerical results show that our approach is robust, competitive and fast.

Fuzzy C-Means Cluster Segmentation Algorithm Based on Bacterial Colony Chemotaxis

2011 ◽  
Vol 421 ◽  
pp. 465-469
Author(s):  
Yan Ling Li ◽  
Gang Li
Keyword(s):  
Image Segmentation ◽  
Optimal Solution ◽  
Segmentation Algorithm ◽  
Bacterial Colony ◽  
Fcm Algorithm ◽  
Fuzzy C Means ◽  
Robust Segmentation ◽  
Initial Cluster ◽  
Local Optimal Solution ◽  
Stochastic Selection

Fuzzy C-Means(FCM) algorithm is one of the most popular methods for image segmentation, but it is in essence a technology of searching local optimal solution. The algorithm’s initial clustering centers are stochastic selection which causes it to depend on the selection of the initial cluster centers excessively. For this reason, fuzzy C-means cluster segmentation algorithm based on bacterial colony chemotaxis (BCC) is proposed in this paper. Firstly, initial cluster centers of FCM algorithm is get by BCC algorithm. Then, the images are segmented using FCM algorithm. Experimental results show that the proposed algorithm used for image segmentation can segment images more effectively and can provide more robust segmentation results.

An Effective Method for Combining Kernels with Class Separability

2012 ◽  
Vol 220-223 ◽  
pp. 2224-2227
Author(s):  
Ting Hua Wang ◽  
Qiong Zhang ◽  
Hai Hui Xie ◽  
Jun Ting Chen
Keyword(s):  
Image Segmentation ◽  
Convex Combination ◽  
Experimental Results ◽  
Evaluation Function ◽  
Data Set ◽  
Kernel Weights ◽  
Class Separability ◽  
Improve Accuracy ◽  
The Individual

We propose a simple but effective method to determine the kernel weights for convex combination of multiple kernels. The key property of the proposed method is that it adopts a class separability criterion as the evaluation function to measure the goodness of the individual kernel. Based on the principle of class separability, we assign a weight to each kernel that is proportional to the quality of the kernel. Experimental results on Image Segmentation data set show the proposed method can improve accuracy in comparison with that using a single kernel or uniformly-combined kernel.

Sign in / Sign up

Export Citation Format

Share Document