Computer vision is one of the most complex and computationally intensive problem. Like any other computationally intensive problems, parallel pro cessing has been suggested as an approach to solving the problems in com puter vision. Computer vision employs algorithms from a wide range of areas such as image and signal processing, advanced mathematics, graph theory, databases and artificial intelligence. Hence, not only are the comput ing requirements for solving vision problems tremendous but they also demand computers that are efficient to solve problems exhibiting vastly dif ferent characteristics. With recent advances in VLSI design technology, Single Instruction Multiple Data (SIMD) massively parallel computers have been proposed and built. However, such architectures have been shown to be useful for solving a very limited subset of the problems in vision. Specifically, algorithms from low level vision that involve computations closely mimicking the architec ture and require simple control and computations are suitable for massively parallel SIMD computers. An Integrated Vision System (IVS) involves com putations from low to high level vision to be executed in a systematic fashion and repeatedly. The interaction between computations and information dependent nature of the computations suggests that architectural require ments for computer vision systems can not be satisfied by massively parallel SIMD computers.

Computer vision is regarded as one of the most complex and computationally intensive problems. An integrated vision system (IVS) is considered to be a system that uses vision algorithms from all levels of processing for a high level application (such as object recognition). A model of computation is presented for parallel processing for an IVS. Using the model, desired features and capabilities of a parallel architecture suitable for IVSs are derived. Then a multiprocessor architecture (called NETRA) is presented. This architecture is highly flexible without the use of complex interconnection schemes. The topology of NETRA is recursively defined and hence is easily scalable from small to large systems. Homogeneity of NETRA permits fault tolerance and graceful degradation under faults. It is a recursively defined tree-type hierarchical architecture where each of the leaf nodes consists of a cluster of processors connected with a programmable crossbar with selective broadcast capability to provide for desired flexibility. A qualitative evaluation of NETRA is presented. Then general schemes are described to map parallel algorithms onto NETRA. Algorithms are classified according to their communication requirements for parallel processing. An extensive analysis of inter-cluster communication strategies in NETRA is presented, and parameters affecting performance of parallel algorithms when mapped on NETRA are discussed. Finally, a methodology to evaluate performance of algorithms on NETRA is described. Choudhary, Alok N. and Patel, Janak H. and Ahuja, Narendra Unspecified Center NAG1-613...

Proceedings -- Parallel Computing.

Over the last twenty years, automation and robotics have played an increasingly important role in a variety of application domains including manufacturing, hazardous environments, defense, and service industries. Space is a unique environment where power, communications, atmospheric, gravitational, and sensing conditions impose harsh constraints on the ability of both man and machines to function productively. In this environment, intelligent automation and robotics are essential complements to the capabilities of humans. In the development of the United States Space Program, robotic manipulation systems have increased in importance as the complexity of space missions has grown. Future missions will require the construction, maintenance, and repair of large structures, such as the space station. This volume presents the effords of several groups that are working on robotic solutions to this problem. Much of the work in this book is related to assembly in space, and especially in-orbit assembly of large truss structures. Many of these so-called truss structures will be assembled in orbit. It is expected that robot manipulators will be used exclusively, or at least provide partial assistance to humans. Intelligent Robotic Systems for Space Exploration provides detailed algorithms and analysis for assembly of truss structure in space. It reports on actual implementations to date done at NASA's Langley Research Center. The Johnson Space Center, and the Jet Propulsion Laboratory. Other implementations and research done at Rensselaer are also reported. Analysis of robot control problems that are unique to a zero-gravity environment are presented.