Pipeline Architecture and Parallel Computation-Based Real-Time Stereovision Tracking System for Surgical Navigation

This paper studies the development of a real-time stereovision system to track multiple infrared markers attached to a surgical instrument. Multiple stages of pipeline in field-programmable gate array (FPGA) are developed to recognize the targets in both left and right image planes and to give each target a unique label. The pipeline architecture includes a smoothing filter, an adaptive threshold module, a connected component labeling operation, and a centroid extraction process. A parallel distortion correction method is proposed and implemented in a dual-core DSP. A suitable kinematic model is established for the moving targets, and a novel set of parallel and interactive computation mechanisms is proposed to position and track the targets, which are carried out by a cross-computation method in a dual-core DSP. The proposed tracking system can track the 3-D coordinate, velocity, and acceleration of four infrared markers with a delay of 9.18 ms. Furthermore, it is capable of tracking a maximum of 110 infrared markers without frame dropping at a frame rate of 60 f/s. The accuracy of the proposed system can reach the scale of 0.37 mm RMS along the x- and y-directions and 0.45 mm RMS along the depth direction (the depth is from 0.8 to 0.45 m). The performance of the proposed system can meet the requirements of applications such as surgical navigation, which needs high real time and accuracy capability.

Pipeline Architecture and Parallel Computation-Based Real-Time Stereovision Tracking System for Surgical Navigation

This paper studies the development of a real-time stereovision system to track multiple infrared markers attached to a surgical instrument. Multiple stages of pipeline in field-programmable gate array (FPGA) are developed to recognize the targets in both left and right image planes and to give each target a unique label. The pipeline architecture includes a smoothing filter, an adaptive threshold module, a connected component labeling operation, and a centroid extraction process. A parallel distortion correction method is proposed and implemented in a dual-core DSP. A suitable kinematic model is established for the moving targets, and a novel set of parallel and interactive computation mechanisms is proposed to position and track the targets, which are carried out by a cross-computation method in a dual-core DSP. The proposed tracking system can track the 3-D coordinate, velocity, and acceleration of four infrared markers with a delay of 9.18 ms. Furthermore, it is capable of tracking a maximum of 110 infrared markers without frame dropping at a frame rate of 60 f/s. The accuracy of the proposed system can reach the scale of 0.37 mm RMS along the x- and y-directions and 0.45 mm RMS along the depth direction (the depth is from 0.8 to 0.45 m). The performance of the proposed system can meet the requirements of applications such as surgical navigation, which needs high real time and accuracy capability.

Web Services Workflows—Composition, Co-Ordination, and Transactions in Service-Oriented Computing

Web services can be seen as a newly emerging research area for Service-oriented Computing and their implementation in Service-oriented Architectures. Web services are self-contained, self-describing modular applications or components providing services. Web services may be dynamically aggregated, composed, and enacted as Web services Workflows. This requires frameworks and interaction protocols for their co-ordination and transaction support. In a Service-oriented Computing setting, transactions are more complex, involve multiple parties (roles), span many organizations, and may be long-running, consisting of a highly decentralized service partner and performed by autonomous entities. A Service-oriented Transaction Model has to provide comprehensive support for long-running propositions including negotiations, conversations, commitments, contracts, tracking, payments, and exception handling. Current transaction models and mechanisms including their protocols and primitives do not sufficiently cater for quality-aware and long running transactions comprising loosely-coupled (federated) service partners and resources. Web services transactions require co-ordination behavior provided by a traditional transaction mechanism to control the operations and outcome of an application. Furthermore, Web services transactions require the capability to handle the co-ordination of processing outcomes or results from multiple services in a more flexible manner. This requires more relaxed forms of transactions—those that do not strictly have to abide by the ACID properties—such as loosely-coupled collaboration and workflows. Furthermore, there is a need to group Web services into applications that require some form of correlation, but do not necessarily require transactional behavior. The purpose of this paper is to provide a state-of-the-art review and overview of some proposed standards surrounding Web services composition, co-ordination, and transaction. In particular the Business Process Execution Language for Web services (BPEL4WS), its co-ordination, and transaction frameworks (WS-Co-ordination and WS-Transaction) are discussed.

面向高性能数值计算的并行计算模型DRAM(h)

提出了一个基于存储层次的新并行计算模型DRAM（h），并在该模型下对两个经典并行数值计算算法的不同实现形式：四种形式并行了三角方程求解（PTRS）和六种形式无列选主元并行LU分解（PLU），进行了分析。模型分析表明，具有近乎相同时间和空间复杂性的同一算法不同实现形式在该模型下会有完全不同的存储复杂度。作者在日立公司SR2201 MPP并行机、曙光3000超级服务器和中国科学院科学与工程计算国家重点实验室（LSEC）的128节点Linux Cluster等三种并行计算平台上对模型分析结果进行了实验验证。结果表明，该模型分析在绝大数情况下都能较好地与不同实验枰台上的实验结果吻合。个别出现偏差的分析结果，在根据计算平台的存储层次特点修改模型分析的假定后，也能够进行解释。这说明了该模型以不同形式的算法实现进行存储访问模式区分的有效性。对在计算模型中加入指令/线程级并行的可行性和方法的研究是下一步的工作。