Advanced Materials Research Vols. 341-342

Abstract: This paper describes the techniques of the processing of automatic segmention vertebra from MRI image. The final choice is use regional growth to segment the vertebra automatically. When processing the image with algorithm of regional growth, use the Contour-tracing method to deal with the region of spine in order to extract contour. Then use mathematical morphology on the experimental results as post-processing.

Abstract: The aeolian vibration of overhead transmission lines frequently brings about the breaking fatigue of wires, which makes a serious threat to the security of transmission lines.Today, more and more attentions have been paid to the aeolian vibration both in research and commercial realms . As aeolian vibration monitoring can accurately monitor wire damage, it is conducive to timely maintenance and avoiding accidents. In our paper, we propose a system model adopting bending amplitude method to estimate bending amplitude, calculate dynamic bending strain of wire and obtain early warning tips by comparing with strain chart based on the research of principles of aeolian vibration. Besides, A prototype system is developed to realize the effectiveness of the proposed model.

Abstract: As information technology is becoming more and more popular, infrared ray technology also has been rapidly increasing especially after A.D. 2000. Infrared ray detection and control technology has been widely applied in various domains of the national economy and people’s daily life. The paper makes analysis on the characteristics of pyroelectric infrared sensor output signal, and puts forward a high precision pyroelectric infrared acquisition circuit. This circuit has high sensitivity, high accuracy, low noise, anti-interference and other features, and it is fit for all sorts of practical detection controlling circuits.

Abstract: Aiming at the LVDS high speed data communication used on long distance, this paper designed the method of data reading and status monitoring of the recorder during the experiment. The transferring cables are twisted-pair, its transferring distance is 100m and the data transmission speed is 12Mbytes/s. With the method introduced in this paper, great deal of time is saved during the data retrieving process. It also increases the efficiency of the experiment, even better, the transmission medium is cheaper, more simply, and easier to be achieved than the optical fiber and coaxial cables.

Abstract: Relative stability of the internal environment is the basis for the body’s all intelligent activities, and the endocrine system plays an irreplaceable role in maintaining that stability. Based on the self-organization mechanism of the hormone reaction diffusion in the endocrine system, this paper presents the artificial endocrine network model and the model-based learning algorithm. The model depends on the diffusion of artificial hormones and its reaction with suitable receptors to achieve the dynamic balance control of the artificial endocrine network. In order to validate the feasibility of the model and algorithm, this paper makes a simulation experiment of robotic navigation control, whose results also show that the model and its algorithm has good adaptive solving ability.

Abstract: Based on the changeability, dynamic and reconfiguration of network collaborative activity in cluster, the frame of network resource planning in cluster based on bill of hierarchies and service-oriented architecture is eatabilished in this paper, which provides the distributed share of network resource in cluster.

Abstract: To improve the efficiencies in kinematics, dynamics analysis and finite element (FE) calculation, distributed computing is used in the project of Chinese mechanical virtual human (CMVH). A three-dimensional (3D) musculoskeletal model of a male human and its finite element model have been constructed according to the male dataset of Chinese visible human (CVH). Many servers and software have been architected by using the method of distributed computing. Finally, a distributed computing platform by using these models to solving the parameters has been established. This distributed computing platform will provide wide applications in the areas such as medical engineering, robot design, physical and art education, sport, ergonomics and traffic accident analysis etc.

Abstract: Flash-based solid state disks (SSD) is a performance based data storage technology that optimizes the use of flash-based technology to implement its data storage capabilities compared with mechanically available data storage technologies. It has been argued in theory and practice that SSD devices are better performers compared with mechanical devices. To improve the efficiency of a flash memory SSD device, it is important for it to be designed to be computationally support parallel operations.

Abstract: With the rapid development of computer network, the network architecture is increasingly complicated; the scale of network extends continuously. Therefore, the functions and positions of the network management are also more and more important. The Web-Based network management, making users manage the network and system expediently and simply, combining Web technology and modern distributed network management technology, which can also be updated and customized easily to meet the different needs, is an important direction of the development of the network management technology in the future. This paper proposes an a SNMP-based Web network management system using B/S structure, which has features of flexible structure, easily extensible, user-friendly features and so on.

Abstract: This paper discusses the uncertainty in remote sensing image segmentation by studying the uncertainty in cloud model , qualitative to quantitative conversion, proposed one way to study the image segmentation based on the combination of Parallel mechanism and cloud mode. Proved by experiments, this method can better describe the uncertainty of image target. It can quickly and accurately segment the target, the efficiency is far superior then the traditional image segmentation algorithms.