In the field of indefinite spatial data reconstruction, Multiple-Point Statistics (MPS) has been widely used, but its applicability is affected due to the high computational cost. A spatial data reconstruction method based on a multi-resolution Generative Adversarial Network (GAN) model was proposed by using a pyramid structured fully convolutional GAN model to learn the data training images with different resolutions. In the method, the detailed features were captured from high-resolution training images and large-scale features were captured from low-resolution training images. Therefore, the image reconstructed by this method contained the global and local structural information of the training image while maintaining a certain degree of randomness. By comparing the proposed algorithm with the representative algorithms in MPS and the GAN method applied in spatial data reconstruction, it can be seen that the total time of 10 reconstructions of the proposed algorithm is reduced by about 1 h, the difference between the average porosity of the algorithm and the training image porosity is reduced to 0.000 2, and the variogram curve and the Multi-Point Connectivity (MPC) curve of the algorithm are closer to those of the training image, showing that the proposed algorithm has better reconstruction quality.

It has been widely concerned to apply fractional partial differential equations in image processing, especially in the image denoising and image Super Resolution (SR) reconstruction. The current research results have shown the advantages and effects of fractional order applications. The theory and model of fractional partial differential equations in image denoising and image super-resolution reconstruction were introduced and discussed. The simulation results show that the methods based on fractional partial differential equations has more advantages than the methods based on integer order partial differential equations in terms of denoising and reducing the staircase effect. Finally, the related research problems were pointed out.

College enrollment consultation service usually relies on artificial reply or keyword matching Question and Answer (Q&A) system, which exists the problems of low efficiency and irrelevant answers. In addition, a consultation text is often a short statement, therefore its vectorized representation may easily lead to the high-dimensional sparse problem. To solve the problems mentioned above, an enrollment consultation algorithm based on Stacked Denoising Sparse AutoEncoders (SDSAE) was proposed. First of all, to improve generalization ability of the algorithm, an autoencoder network was used to extract features and reduce the data dimension; at the same time, dataset enhancement technique and noise-adding technique were introduced to solve the problems of small training sample set and uneven classification. After low dimensional representation of short texts being obtained, a text classification was conducted afterwards by using Back Propagation (BP) algorithm. The experimental results show that the proposed algorithm has a better classification performance over BP, Support Vector Machine (SVM), Extreme Learning Machine (ELM) algorithm and etc., and it significantly improves the classification effect of enrollment consultant texts.

In an Executable and Linkable Format (ELF) executable program, Global Offset Table (GOT) was used to store the absolute addresses of library functions. But in Linux operation system, GOT dereference and GOT overwrite are two common vulnerability exploit methods. Through analyzing the GOT feature, a protection method for GOT based on address randomization and segment isolation was proposed and implemented. With modifying the ELF loader program, all sections which pointed to the GOT were loaded into random memory addresses. Using segment isolation technology, all instructions with reference to GOT used a new segment register. The experimental results prove that the proposed method can not only defense against the exploit method of GOT effectively, but also has a very low cost of average 2.9 milliseconds.

Route choice is a practical problem that cannot be avoided in daily life. Since pedestrians still need to identify if the real landmark is the same landmark in route instructions under the assistant of pedestrian navigation system, a model of pedestrian road network was established. A pedestrian route choice model was proposed by introducing the subjective judgment of delay time and distance into Prospect Theory (PT). The simulation experiment is performed, in this study, on partial region of China University of Geosciences (Wuhan), the results show that the subjective judgment delay time in the same Origin-Destination (OD) pair calculated by the proposed pedestrian route choice model is 0.6 s less than the shortest subjective judgment delay time, and the lengths calculated by it are longer than the shortest routes, however they are all less than 16 m. The experiment results show that the proposed model gives a more realistic route, which meets the actual needs of pedestrians travel.

Concerning the limitations that the accuracy of prediction is low and the classification on box-office is not significant in application, this paper proposed a new model to predict box-revenue of movie, based on the movie market in reality. The algorithm could be summarized as follows. Firstly, the factors that affected the box and format of the output were determined. Secondly, these factors should be analyzed and quantified within [0, 1]. Then, the number of neurons was also determined, aiming to build up the architecture of the neural network according to input and output. The algorithm and procedure were improved before finishing the prediction model. Finally, the model was trained with denoised historical movie data, and the output of model was optimized to dispel the randomness so that the result could reflect box more reliably. The experimental results demonstrate that the model based on back propagation neural network algorithm performs better on prediction and classification (For the first five weeks, the average relative error is 43.2% while the average accuracy rate achieves 93.69%), so that it can provide a more comprehensive and reliable suggestion for publicity and risk assessment before the movie is on, which possesses a better application value and research prospect in the prediction field.

In this paper, a new image encryption algorithm was presented based on the fractional-order Chen chaotic system, for fractional-order chaotic dynamical systems have more complex dynamical behaviors than those of integer-order systems and can provide more freedom for image encryption schemes. In the transmitter, the positions of the image pixels were scrambled by the chaotic signal generated by the driving system firstly. Then the disturbed image was embedded into the chaotic signal and the encrypted image for transmission was obtained. In the receiver, the chaotic signal was removed by the synchronization system. Then the inverse process of pixel scrambling was carried out and the original image was recovered. The security of the proposed algorithm was analyzed in the end. The experimental results demonstrate that the encryption algorithm is of high security and has good research value and application prospects.

To improve the resource utilization, the task resource requirement features of processor, network, disk and so on were efficiently estimated through analyzing the way of the task request, measuring the performance of application program in task or simulating to run the task. Afterwards, according to the features of resource requirement, the tasks could be classified into processor bound task, communicate bound task, disk bound task and others. And then the classified tasks were integrated with specific virtual machines to make all kinds of resources to be used efficiently. The research shows that the method can classify the task efficiently. And compared to unclassified method, it can reduce the times of virtual machines migration or integration.

The application of traditional software prediction model has poor generalized performance. This paper put forward a software reliability prediction model based on Learning Vector Quantization (LVQ) neural network. First, this paper analyzed the structure characteristics of LVQ neural network and its relation with software reliability prediction. Then the network was used to predict the software reliability. In the end, the authors confirmed the algorithm through multiple simulation experiments under the Matlab environment and the data from Metrics Data Program (MDP) database of National Aeronautics and Space Administration （NASA） of USA. The experimental results indicate that the method is feasible and has a higher prediction precision than the traditional software prediction method.