Search

In order to solve the problems of inadequate global search and easy to fall into local extremum, this paper proposed a multi-strategy fusion improved adaptive tunicate swarm algorithm（MITSA）. Firstly, this paper used the best point set strategy to improve population diversity during initialization. Secondly, it offered a multi-elite cooperative guidance mechanism to optimize the tunicates’ location information, and enhance the global exploration ability of the algorithm by increasing the exploration possibility of unknown search area. Then, the algorithm achieved dynamic global and local search balance by introducing adaptive weight factors into the group behavior stage. Meanwhile, probabilistic wavelet variation strategy promoted individual dynamic fine-tuning to enhance the anti-stagnation ability of the algorithm, while using the greedy principle to retain excellent information to help the population to the food source. Finally, this paper proved the global convergence of the improved algorithm based on Markov chain theory. Through the numerical simulation of the benchmark test function and CEC2014 complex function, the experimental results and Wilcoxon rank sum test results comprehensively verify MITSA’s excellent convergence accuracy, robust robustness and high-dimensional scalability. [ABSTRACT FROM AUTHOR]

Aiming at the high mismatching rate of traditional local stereo matching algorithm at the depth discontinuities, this paper proposed an occlusion information stereo matching approach based on adaptive weight. Firstly, left-right consistency check algorithm detected the occlusion area of the reference image and the target image . Then, this paper utilized occlusion information to reduce the weight of pixels in the occlusion area at the process of cost aggregation, and scanline propagation filled occlusion pixels with horizontally nearest points in the disparity refinement stage. Finally, it calculated the mismatching rate of disparity map according to the Middlebury standard image dataset. The experimental results show that compared with adaptive support weight approach, the proposed approach decreases the mismatching rate by 16%. Meanwhile, it not only has lower mismatching rate in the depth discontinuous region, but also improves the accuracy of algorithm matching. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the WOA was easy to fall into local optimum and low convergence precision, this paper proposed an improved whale optimization algorithm (IWOA). The algorithm initialized the population by quasi-reverse learning methods and improved the diversity of the population. Then the algorithm modified the linear convergence factor to a nonlinear convergence factor, which was beneficial to balance the global search ability and local development ability. In addition, the algorithm improved the local search ability of the whale optimization algorithm by increasing the adaptive weight and imp roved convergence precision. Finally, the algorithm adjusted the whale optimization algorithm in time by a random differential mutation strategy to avoid falling into the local optimum. It selected nine benchmark functions in the experiment, and iterated all the algorithms 30 times. The improved whale optimization algorithm compared to the original whale optimization algorithm and five improved whale optimization algorithms, the results show that the mean and standard deviation of the algorithm are better than other algorithms, the convergence curve of the algorithm is also superior to most other algorithms. It shows that the imp roved whale optimization algorithm has the best convergence accuracy and algorithm stability, and the convergence speed is significantly faster than most other improved whale optimization algorithms. [ABSTRACT FROM AUTHOR]

Industrial processes often contain significant strong nonlinearity and time-varying behavior. Traditional ElM based soft sensor sometimes fails to make use of data information effectively and has poor prediction performance. This paper proposed an improved particle swarm optimization algorithm, which had better convergence speed and search ability than standard PSO. This algorithm used the characteristics of Gaussian distribution to update the inertia weight adaptively and changed learning factor linearly. It optimized the penalty coefficient and kernel parameter of ELM to obtain a group of optimal parameters. This algorithm was applied to soft sensor modeling for the debutanizer column process. The simulation results verify that the proposed method has good prediction accuracy and generalization performance. [ABSTRACT FROM AUTHOR]

In order to improve the retrieval efficiency when facing a large number of lung images,this paper presented a retrieval method for lung nodules CT image based on supervised hashing. The method firstly built lung nodules image database through image pre-procession and extracted the multi-features from gray,morphology and texture. Then,it utilized supervised information to construct hash functions and translated the multi-features into short hash codes. Finally,it retrieved the similar lung nodules CT images according to similarity calculation with adaptive weight. Experimental results show that the proposed method can effectively achieve the rapid retrieval of lung nodules CT image and the classification of benign and malignant tumor can reach 89. 45%. [ABSTRACT FROM AUTHOR]