In view of the drawbacks of the standard Discrete Particle Swarm Optimization (DPSO) algorithm such as slow convergence speed and easily trapping into local optima, an Improved Local-search-based Chaotic Discrete Particle Swarm Optimization (ILCDPSO) algorithm based on excellence coefficient was proposed and then applied to Traveling Salesman Problem (TSP). In this algorithm, each edge was assigned an appropriate excellence coefficient based on the principle of roulette selection. This helped to improve the selection probability of short edge, thus improving the optimization ability and convergence speed of the algorithm. In order to further improve the accuracy of solution, a local search strategy was employed such that the exploration ability of the algorithm could be improved by adjusting the routes of cities in the given neighborhood for each city. Moreover, a chaotic sequence was integrated into the iteration formula to enhance the randomness and diversity of particles and hence increasing the global searching ability of the proposed algorithm. Finally the algorithm was evaluated by some typical instances in the internationally commonly used library of TSP (TSPLIB) and compared with Particle Swarm Optimization (PSO) algorithm, Improved Particle Swarm Optimization (IPSO) algorithm, and Chaotic Particle Swarm Optimization (CPSO) algorithm, etc. The experiment data show that, under the same experimental conditions, ILCDPSO can achieve optimal solutions with less average iterations than other algorithms and has the highest ratio of number for obtaining optimal solutions. The research results indicate that ILCDPSO algorithm performs better than other algorithms in terms of convergence speed, global optimization ability and stability, and it is a comparatively potential intelligent algorithm for solving TSP.

To overcome the problem of easily trapping into local optima of standard Artificial Bee Colony (ABC) algorithm, the roulette selection strategy of ABC was modified and an improved ABC based on dynamic evaluation selection strategy (DSABC) algorithm was proposed. Firstly, the quality of each food source position was evaluated dynamically according to the times that the food source position had been continuously updated or stagnated within a certain number of iterations so far. Then, onlooker bees were recruited for the food source according to the obtained value of the evaluation function. The experimental results on six benchmark functions show that, compared with standard ABC algorithm, the proposed dynamic evaluation selection strategy modifies the selection strategy of ABC algorithm, and greatly improves the quality of solution of DSABC algorithm, especially for function Rosenbrock with different dimensions, the absolute error of the best solution reduces from 0.0017 and 0.0013 to 0.000049 and 0.000057, respectively; Moreover, DSABC algorithm can avoid the premature convergence caused by the decrease of population diversity at later stage and improve the accuracy and stability of solutions, thus provides an efficient and reliable solution method for function optimization.

With reference to the properties of multiscale and shift invariance of nonsubsampled Contourlet transform, and concerning the characteristics of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) images, a medical images fusion method was proposed．The proposed method fused the low frequency subband and high frequency subband of these medical images separately by the regional feature strategy. The paper introduced the judgment criteria of images fusion and expatiated on the principle and implementation of Nonsubsampled Contourlet Transform (NSCT). And this gave the subjective judgment and numeric measurement of the fusion images based on visual effect and information indexes. To evaluate the performance of the proposed algorithm, the authors compared the results with those of the algorithms, such as wavelet transform and Contourlet transform. The CT and MRI images simulation results of mandibular system indicate that the proposed method outperforms the others in terms of both visual quality and objective evaluation criteria, while it can integrate and maintain much more effective and detailed information as well.

Concerning the faults of local extremum in image registration based on mutual information, a new medical image registration method based on Powell and improved genetic algorithm was proposed in this paper. It put forward an improved method regarding the shortcomings of the standard genetic algorithm, such as slow convergence and prematurity that will result in artifacts, and generated the iteration individual by Logistic chaos map. This method utilized the multi-resolution analysis strategy and searched for the optimal of the objective function by this hybrid optimized algorithm in the lowest resolution image level. Then it continued the optimization course and accomplished the image registration by this optimal data with the Powell algorithm. The experimental results indicate that this algorithm can effectively improve the image registration velocity and avoid local extremum of the operator while getting better performance of image precision in contrast to the Powell algorithm and unimproved genetic algorithm.

With the demand of real-time and dynamic scalability processing for multi-source mass data in transportation, this paper proposed a distributed in-memory database experimental platform based on GemFire. The platform used the attributes of GemFire, such as key-value data storage structure and distributed dynamic membership. The actual data from floating car system was used to complete the performance analysis in cloud computing architecture. The experimental results show that the platform can shorten the calculation time of mass data to less than 10% of the existing system and basically satisfy the application requirements of transport data resources integration in cloud computing platform.

It is essential to classify E-commerce customers accurately and take suitable marketing strategy for the development of E-commerce. Based on the RFM (Recency,Frequency,Monetary) model used in customer classification of traditional retail, the total profit attributes were introduced to set RFP model and K-Means clustering method was used to analyze customers of a firm in E-commerce and compared with the RFM model. The relevance of model attributes impact on clustering results was analyzed. Finally, the paper gets six conclusions about the comparison between the both models and puts forward four strategies for E-Commerce and other sales industries.

Model checking algorithm was investigated for Interactive Markov Chains (IMC). IMC is a concurrent system model, combining functional feature and performance properties. Model checking is one kind of automatic function confirmation and the performance evaluation technology. The presented algorithm is a combination of traditional function confirmation and performance evaluation, and which is in consistency with the original algorithm. Experimental results have shown the algorithm can achieve a good performance and can be used to evaluate for large and complex systems.