Advanced Materials Research Vols. 546-547

Abstract: This paper puts forward a survey on the existing 3D medical image visualization methods. These methods are classified into two groups and typical algorithms in each group are described and analyzed. At last future research directions are discussed.

Abstract: With the rapid development of cloud computing, the datacenters are widely used in various areas. Connected by optical fiber, the datacenters are always distributed unbalanced, as the service requirements are very different in different regions. To ensure each datacenter’s stable operation and make them share the equal amounts of risk, the service resources need to be distributed to distinct datacenters proportionately. Considering the service recourses of datacenters, this paper is aimed at proposing a dynamic load balancing algorithm. The algorithm, named Average Variance, can help the new tasks to be distributed with a dynamic adjustment on the basis of the current datacenter’s situation in real-time. And in this way, it can achieve the minimum average variance among datacenter’s occupation of service resource. The simulation results demonstrate that the Average Variance has a better performance than the Round Robin. Furthermore, the main consideration of the node in this paper is the utilization in each CPU.

Abstract: To offer personalized recommendation service to web users, it adopts improved FP-Growth algorithm, introduces its implementation methods and directly applies it to the recessive knowledge mining of web site information category in details. Via mining the website data and analysing association rules, useful relavant knowledge is obtained, tendancy of website visiting can be prediced and personalized service will be prescribed which make website more friendly and satisfactory.

Abstract: Image processing techniques in virtual space are described in the paper. The images are merged in the wavelet domain to acquire required scene, and dealt with perspective correction. High-quality program can be obtained by stitching a set of images which are processed. And through VC++ experiments, feasibility of the algorithm is proved.

Abstract: The knowledge representation of the traditional artificial intelligence used different modeling methods and the different development tools, it led to the lack of interoperability between all kinds of knowledge, ontology solved the problem. Ontology, which is a model in semantic and knowledge hierarchy describing the concept and the relationship between the concepts, has been the focus of the field of artificial intelligence since it was proposed. This paper explored the knowledge representation based on ontology in the field of artificial intelligence, built the maize domain knowledge ontology, the result shows: ontology can effectively solve the heterogeneous problem of expression of complex knowledge, makes the computer to understand information for the semantic level, and benefit to develop the intelligent systems of maize.

Abstract: Most conventional tracking gate algorithms only use the targets’ kinematic measurement information, which is typically resulted in great uncertainties of measurement-to-track association for multi-target tracking in clutter. The problem of constructing tracking gates using targets' class information is considered. The proposed algorithm integrates targets' identity information into the traditional tracking gating techniques. First, a class-dependent gate corresponding to each class of targets is developed. Second, the algorithm for constructing the class-dependent gate is given. Simulations are carried out to examine the proposed algorithm, where the simulation scenario shows that the measurement-to-track association using the class-dependent gating algorithm is significantly better than traditional method.

Abstract: As the amount of sales increases rapidly, amount of data become very huge, and the management of customers’ relationship also becomes a more complex problem. Using data mining to analyze data to discover the rules and knowledge among them so that customized services in electronic commerce could be sustained and enterprises’ sales could be more intelligential. With data mining, we can do the following things. Firstly, it analyzes the customers’ shopping behavior and preference with association rules so that it can provide the recommending in the shopping process to make customers getting the right goods more convenient and faster. Secondly, it uses decision tree to classify the customers so that it makes a better communication between customers, provides customized shopping user interface to them and gives the pertinent advertisement to them.

Abstract: A current problem in the development of passive sonar is the classification of different noise sources. The existing Automatic Underwater Target Recognition Technique (AURT) is mainly based on spectrum analysis of radiated noise in passive sonar. However, with the development of noise control techniques, the radiated noise of underwater targets have been reduced enormously in the past few years, even the working states of the same target also weaken the stationarity of spectrum features of its radiated noise. The objective of this paper concentrates on the AURT with multi-method analysis on radiated noise and decision-makings fusion. A kind of multi-classifier decision-makings fusion model to overcome the non-stationarity is presented. Its application of the model on the data derived from sea trial confirms its validation even in the case of low SNR, and the classification rate is above 83%, better than that of spectrum analysis only.

Abstract: Watershed is an image segmentation algorithm based on mathematical morphology, which can determine the boundary of connected section efficiently and effectively. But the traditional watershed algorithm is sensitive to noise. To overcome the weakness of classical watershed, this paper presents an improved watershed algorithm based on gradient transform, open-close reconstruction and distance transform. The experiment result shows that application of this improved watershed algorithm in cell image segmentation has a good performance.

Abstract: Coarse-Grained Reconfigurable Architectures (CGRA) have proved to be the potential candidates to meet the high performance, low power and flexibility required by embedded systems. In this paper, we implemented a High Profile Intra Predictor for H.264/AVC decoder on a novel coarse-grained reconfigurable processor- Remus (Reconfigurable Multi-media System). We proposed the pipeline and parallel scheduling process for intra prediction algorithm and the simulation results show that 548 clock cycles are consumed for the worst case of intra macro blocks.