Since the dark channel prior transmission and atmospheric light in the bright region are poorly estimated, an improved dehazing algorithm combined with atmospheric light and transmission was proposed. On the basis of analysis of the characteristics of Gaussian function, a preliminary transmission was estimated through the Gaussian function of dark channel prior of a fog image, and the maximum and minimum operations were used to eliminate the block effect. Next, the atmospheric light was obtained by atmospheric light description area, which was acquired by halo operation and morphological dilation operation. Finally, a clear image could be reconstructed according to the atmospheric scattering model. The experimental results show that the proposed algorithm can effectively remove the fog from the image and the recovered effect of thick fog is better than the comparison algorithms, such as dark channel prior, meanwhile the algorithm has a faster processing speed and is suitable for real-time applications.

The extraction of multi-view matching points is one of the key problems in 3D reconstruction of multi-view image scenes, and the accuracy of the reconstruction will be affected directly by the extraction result. The extraction problem of multi-view matching points was converted into the dynamic connectivity problem and a Union-Find (UF) method was designed. The nodes of UF were organized using efficient tree structure in which parent-link connection was adopted. In the process of increasing the matching points, only the addressing parameters of a single node needed to be modified, which avoided comparing the calculation process of the addressing parameters through traversing the array, and the efficiencies of locating and modifying were improved. A weighting strategy was used to optimize the algorithm, and the weighted encoding method was used to replace the conventional hard encoding, which could balance the structure of the tree and reduce the average depth of the dendrogram. The experimental results of multiple image sets show that the proposed algorithm based on UF can extract more multi-view matching points, and is more efficient than the conventional Breadth-First-Search (BFS) algorithm.

To solve the storage problem of massive information for new generation blog system, combined with the cloud storage, a new blog system architecture named BlogCloud was proposed. By using the distributed storage technology, BlogCloud avoided the bottleneck problem in centralized storage, and had high scalability. With semi-distributed Peer-to-Peer (P2P) architecture, it located storage resources quickly. To avoid the network flipping caused by the instable nodes, only the stable nodes were regarded as storage nodes. The rule of nearest storage was adopted and the data was kept in the client cache to reduce the network transmission. Users were allowed to customize the block size of file, it meant that the large files could be devided into blocks and transmitted parallel to enhance the transmission speed, while the small files did not need to be devided, which saved the cost for segmentation and combination. It also had functions of redundancy to enhance the security and reliability of the data, including storing copies of files in multi-storage nodes and remote backup. The comparison test between BlogCloud and ZSWIN was given on the virtual machine. The results show that the throughput of BlogCloud is higher than ZSWIN obviously; the performance of BlogCloud is degraded when using instable nodes as the storage nodes; BlogCloud has high reliability, it still can run stably when reducing the storage nodes and index nodes. The results verify that BlogCloud can satisfy the storage requirements for the new generation blog system.

To solve the low running speed problem of Knuth39 random number generator, a Knuth39 parallelization method based on Many Integrated Core (MIC) platform was proposed. Firstly, the random number sequence of Knuth39 generator was divided into subsequences by regular interval. Then, the random numbers were generated by every thread from the corresponding subsequence's starting point. Finally, the random number sequences generated by all threads were combined into the final sequence. The experimental results show that the parallelized Knuth39 generator successfully passed 452 tests of TestU01, the results are the same as those of Knuth39 generator without parallelization. Compared with single thread on Central Processing Unit (CPU), the optimal speed-up ratio on MIC platform is 15.69 times. The proposed method improves the running speed of Knuth39 generator effectively, ensures the randomness of the generated sequences, and it is more suitable for high performance computing.

In order to overcome the rendering drawbacks of traditional algorithms that cannot be interacted fluently with the user and have a big time consumption and single rendering result, a ray casting algorithm based on Graphics Process Unit (GPU) was proposed to be used for the real-time volume rendering of medical tomographic images. Different rendering effects can be switched quickly by the proposed algorithm. Firstly, medical tomographic images were read into the computer memory to construct voxels. Afterwards, properties (interpolating, shading and light) of the corresponding voxels were set. The transfer functions of color and opacity were designed to display different organs and tissues. Finally, the volume data were loaded and the ray casting algorithm was executed by GPU. The experiments show that the rendering speed of the proposed algorithm can reach 40 frames per second, which satisfies the clinical application. On the aspect of rendering quality, jags produced in the process of interaction because of resampling on GPU are apparently lower than the ray casting algorithm on CPU. The time consumption of CPU-based ray casting algorithm is about 9 times that of the proposed algorithm.

Aiming at terminal performance bottleneck among current data transfer process, a UDP-based terminal adaptive protocol was proposed. After the analysis and the comparison of many factors which affected terminal performance, this protocol viewed both the previous packet loss ratio and the current one as congestion detection parameters. It employed various rate adaption methods such as finite loop counter and process scheduling function in order to balance performance differences in real-time and ensured reliable and fast data transfer. Compared with traditional idle Automatic Repeat reQuest (ARQ) method, the average delay is reduced by more than 25%. The experimental results show that the proposed algorithm has the features of strong real-time, quick response, and it is compatible with large amount of data transmission, especially suitable for small amount of data transmission in engineering applications.

For single sensor data mining prediction problem of the existence of one-sidedness, proposed the multi-sensor data mining prediction model of combining of information fusion technology and phase-space reconstruction technology. A variety of underground sensors, including gas concentration, wind speed, temperature sensors, are fusion forecasted. To many types of sensor time series data for the study, the first using the method of information fusion, respectively, followed by all kinds of data sensor data level fusion, feature level fusion; Then using the method of correlation integral the integration of two sensor data, respectively, to determine the time delay τ and embedding dimension m two parameters for the reconstruction phase; Finally, combined with the techniques of multivariate phase space reconstruction, fusion phase space the various types of sensor data, using the predictive models based on the weight one-rank local-region of K-Means clustering of multi-sensor data. The data is from the coal mines in Shanxi Province and the New King Wu Yi mine, collection of nearly 20G data to the gas concentration, wind speed, temperature experiment three sensor data, the results show that: For the feature level fusion, the data every 15 minutes period of time after fusion to be effective as a measure of the characteristics of this period, after the prediction model calculations, compared with the time period ,5 minutes, 10 minutes, 20 minutes, the error is minimum ESS=0.003, compared with the current minimum error value of 0.05, the error is greatly decreased, therefore, the integration forecasts’ better, it can more accurately predict the future after 15 minutes of sensor data, people have sufficient time to further provide for the safety assessment of underground basis for making decision.

A concurrent processes grouping output method was proposed to address the problem that storage performance decreases when testing Nucleus for European Modeling of the Ocean (NEMO) system with massive processes on TH-1A supercomputer. This method designed a reasonable control strategy based on TH-1A architecture and split processes into groups to alleviate the process competition of concurrency I/O. Testing result shows that the storage performance of the GYRE012 global sample can be improved more than 33% using the optimization method of concurrent processes grouping output, and in the mean time the total performance can be improved about 28%.

Aiming at ordering objects in ordered information table,a framework based on rough set theory was proposed.An information system was firstly transformed into a binary one which would be simplified by rough set theory.According to the values of the weight function,the objects in the original information tables could be sorted.Experiments illustrate that the framework is feasible and effective.