Applied Mechanics and Materials Vols. 347-350

Abstract: Infrared thermal imager has been widely used in the fields of missile guidance and flaw detection. To identify the target clearly, the advanced one adopts dual bands sensors to capture images. Since of that, there is an urgent need of a fusion of the dual-bands images. The fused result includes much more exhaustive information than any single one, and can better reflect the actual. Among the algorithms used to fuse the dual-band infrared images, the weighted algorithm is the most widely used and easiest to be achieved. Nonetheless, its effect is not desired. We extract the features of the source images and make a fuse based on them on the feature-level. To get a better result, in this paper, the fusion strategy based on the Gradient pyramid transform has been mainly adopted. Meanwhile, there is a comparison with the weighted algorithm. Also, it makes an evaluation and analysis to the experimental data, and finally obtains the desired results.

Abstract: This paper introduces a recursive algorithm of Kalman filter for digital predistorter parameters extraction based on memory polynomials predistorter model. The predistorter model is firstly formulated as linear regression expression. Then we derive the state-space equation of the model and attain the steps of the algorithm. Finally, the accuracy and stability of the algorithm is proved by simulation.

Abstract: Nonlinear feature extraction used standard Kernel Principal Component Analysis (KPCA) method has large memories and high computational complexity in large datasets. A Greedy Kernel Principal Component Analysis (GKPCA) method is applied to reduce training data and deal with the nonlinear feature extraction problem for training data of large data in classification. First, a subset, which approximates to the original training data, is selected from the full training data using the greedy technique of the GKPCA method. Then, the feature extraction model is trained by the subset instead of the full training data. Finally, FCM algorithm classifies feature extraction data of the GKPCA, KPCA and PCA methods, respectively. The simulation results indicate that the feature extraction performance of both the GKPCA, and KPCA methods outperform the PCA method. In addition of retaining the performance of the KPCA method, the GKPCA method reduces computational complexity due to the reduced training set in classification.

Abstract: The cloud storage is the evolution of the cloud computing system, which is based on data storage and management. This thesis is focused on the discussion of the cloud storage model that is built on the Cloud Storage Access Protocol of SNIA, and puts forward the concept of storage pooling at storage management layer with unified virtualization and dynamic management on physical storage devices. The thesis also compares the typical DFS (Distributed File System) on the data management layer in different data conditions.

Abstract: For several special features in the environment of cloud computing, which may be quite different from the centralized computing infrastructure currently available, the existed method of resource allocation used in the grid computing environment may not be suitable for these changes. In our paper, a new allocation algorithm based on Ant Colony Optimization (ACO) is proposed to satisfy the needs of Infrastructure as a Service (IaaS) supported by the cloud computing environment. When started, this algorithm first predicts the capability of the potentially available resource nodes; then, it analyzes some factors such as network qualities and response times to acquire a set of optimal compute nodes; finally, the tasks would be allocated to these suitable nodes. This algorithm has shorter response time and better performance than some of other algorithms which are based on Grid environment when running in the simulate cloud environment. This result is verified by the simulation in the Gridsim environment elaborated in the following section.

Abstract: Wigner-Vill distribution (WVD) will inevitably have cross-terms when it used for time-frequency representation in a multi-component signal. In order to suppress the cross-terms in Wigner-Vill distribution, this paper proposes a joint algorithm based on EMD-ICA. This algorithm resolves a multi-component signal into several IMF components used by EMD at first, and then each IMF component is used FastICA algorithm for processing, and next seeks the Wigner-Vill distribution of each component, finally, add up the results. This method effectively inhibited the emergence of cross-terms in Wigner-Vill distribution, and keeps the properties of time-frequency concentration higher.

Abstract: Image noise removal forms a significant preliminary step in many machine vision tasks, such as object detection and pattern recognition. The original anisotropic diffusion denoising methods based on partial differential equation often suffer the staircase effect and the loss of edge details when the image contains a high level of noise. Because its controlling function is based on gradient, which is sensitive to noise. To alleviate this drawback, a novel anisotropic diffusion algorithm is proposed. Firstly, we present a new controlling function based on Laplacian kernel, then making use of the local analysis of an image, we propose a difference curvature driven to describe the intensity variations in images. Experimental results on several natural and medical images show that the new method has better performance in the staircase alleviation and details preserving than the other anisotropic diffusions.

Abstract: Tree growth management decision-making model can simulate growth management of tree and perform quantitative analysis of tree growth conditions. This paper explores the feasibility of modern information technology in management assessment of tree growth, information technology include neural network, ontology and expert system technology, then ontology technology is used to establish ontology database and knowledge base of tree growth management resource, the growth simulation and tree growth management ontology technology are used to build simulation models of tree growth, then expert systems and neural network technology are combined to simulate tree growth development process of decision-making model. The practice has proved that the research can not only predict the growth conditions of tree and dynamic grasp the growth process of the tree, but also can provide theoretical basis for the analysis and evaluation of tree growth management, greatly improving the level of tree growth management.

Abstract: In order to achieve the quantification analysis and visualization of poplar morphology, the technology framework of poplar simulation based on growth model is made. The paper carries out data collection and field observation, then the morphological structure parameters and knowledge models which can describe different poplar varieties are constructed, in the framework of the poplar growth models we integrate poplar morphological knowledge models and geometric models based on morphological character parameters, make the growth model of poplar can output morphological character parameters and topological structure, and further perform the geometric model reconstruction of poplar growth simulation. Practice has proved that the construction of growth models of poplar can simulate the poplar growth at different growth stages, and provide new ideas and means for growth simulation research of poplar morphology structure.

Abstract: Use genetic algorithm for task allocation and scheduling has get more and more scholars' attention. How to reasonable use of computing resources make the total and average time of complete the task shorter and cost smaller is an important issue. The paper presents a genetic algorithm consider total task completion time, average task completion time and cost constraint. Compared with algorithm that only consider cost constraint (CGA) and adaptive algorithm that only consider total task completion time by the simulation experiment. Experimental results show that this algorithm is a more effective task scheduling algorithm in the cloud computing environment.