Due to the inadequacies of the speed of star detection and centroid calculation (SDCC), the applications of high-resolution and high-frame-rate image sensors are greatly limited in star trackers, restricting the perfo...
详细信息
Due to the inadequacies of the speed of star detection and centroid calculation (SDCC), the applications of high-resolution and high-frame-rate image sensors are greatly limited in star trackers, restricting the performance of the star trackers. To resolve this problem, a novel distributedparallel super-block-based SDCC method is proposed in this paper. In contrast to the previous SDCC methods, the proposed method divides the star images into several sub-images and processes these sub-images in parallel. Then, by using a super-block-based method and a distributed shared memory-based method to process the stars in sub-images (sub-stars) and in adjacent sub-images (boundary stars), respectively, the position information of all the stars are detected during the parallel scan of the sub-images. The proposed method exhibits a speed that is up to 4 x M times faster than the previous SDCC methods in processing the star images that are divided into M sub-images. Experimental results demonstrate that the proposed method is correct and effective. It is applicable to the use of most high-performance image sensors and resolves the performance limitation of star trackers to achieve better attitude accuracy and attitude update rate.
distributed computing provides a cost-effective solution for computation intensive problems. With the emerging of networking operating system for personal computer (PC), such as WindowsNT, it is now feasible to develo...
详细信息
ISBN:
(纸本)0819425885
distributed computing provides a cost-effective solution for computation intensive problems. With the emerging of networking operating system for personal computer (PC), such as WindowsNT, it is now feasible to develop distributed computing on a network of PCs. In addition, the computing power delivered by a PC is kept increasing whilst the cost is decreasing. Implying that the performance/cost factor for a PC is high and tile computing power delivered by the network is enormous. In this paper, we describe a software system which enables users to develop distributed computing program using the SPMD (Single Program Multiple Data) paradigm very quickly under the WindowsNT operating system. The programming model for the system is simple and a user can control the system through a graphical interface. The results show that our system provides a reasonable speedup in solving imageprocessing problems.
The paper presents a data and task parallel low-level imageprocessing environment for distributed memory systems. imageprocessing operators are parallelized by data decomposition using algorithmic skeletons. At the ...
详细信息
ISBN:
(纸本)0769512607
The paper presents a data and task parallel low-level imageprocessing environment for distributed memory systems. imageprocessing operators are parallelized by data decomposition using algorithmic skeletons. At the application level we use task decomposition, based on the image Application Task Graph. In this way, an imageprocessing application can be parallelized both by data and task decomposition, and thus beter speed-ups can be obtained. We validate our method on the multi-baseline stereo vision application.
processing a large-scale Synthetic Aperture Radar (SAR) image dataset on a distributed computing infrastructure poses a challenging problem. Large-scale load distribution strategies like multi-installment scheduling (...
详细信息
processing a large-scale Synthetic Aperture Radar (SAR) image dataset on a distributed computing infrastructure poses a challenging problem. Large-scale load distribution strategies like multi-installment scheduling (MIS) assume that the size of the result is negligible compared to the input workloads and hence ignore it in their design. Similarly, numerical methods like particle swarm optimization and their variants are not practical for real-time applications, given their run-time complexities. As both the results retrieval and completion time are crucial for SAR image data processing, in this article, we attempt to provide a thorough theoretical analysis of an adaptive MIS that includes the result retrieval phase. We use the periodic nature of the internal installments to keep the strategy simple and fine-tune the last installment to avoid any idle times in the processors. We derive a closed-form solution for the load fractions and hence, the overall processing time, schedule feasibility criteria, and certain other properties that lead to adaptive scheduling. Finally, we validate our theoretical findings through rigorous simulation studies using a loosely connected virtual machines (VMs) topology for the SAR dataset.
parallelprocessing techniques have been widely promoted as a feasible, and even promising, solution to the very high computational costs associated with practical h-p adaptive finite-element analysis software tools f...
详细信息
parallelprocessing techniques have been widely promoted as a feasible, and even promising, solution to the very high computational costs associated with practical h-p adaptive finite-element analysis software tools for electromagnetic device design and system performance simulation. A combination of emulated and empirical studies designed to explore the validity of these claims are presented and compared. Practical parallelprocessing efficiency results, computed for Sun E450 and v880 parallel workstation platforms (four processors each), are reported.
Increasing acceptance of the necessity for high-order parallelism in order to progress digital processing still leaves open the large question of what machine architectures are best for which class of problem. To help...
详细信息
Increasing acceptance of the necessity for high-order parallelism in order to progress digital processing still leaves open the large question of what machine architectures are best for which class of problem. To help answer this, we are investigating and comparing the use of both SIMD and MIMD architectures for programmable processing in real-time systems. A distributed array machine, Mil-DAP (derived from the original ICL DAP) has been developed and benchmarked on radar, imageprocessing, and on terrain modelling problems. Multi-transputer arrays have been applied to an overlapping set of problems in imageprocessing, FFT and terrain-based computation. The results are compared and preliminary conclusions drawn.
This article presents a new generation in parallelprocessing architecture for real-time imageprocessing. The approach is implemented in a real time image processor chip, called the Xium(TM)-2, based on combining a f...
详细信息
ISBN:
(纸本)0819425885
This article presents a new generation in parallelprocessing architecture for real-time imageprocessing. The approach is implemented in a real time image processor chip, called the Xium(TM)-2, based on combining a fully associative array which provides the parallel engine with a serial RISC core on the same die. The architecture is fully programmable and can be programmed to implement a wide range of color imageprocessing, computer vision and media processing functions in real time. The associative part of the chip is based on patented pending methodology of Associative Computing Ltd. (ACL), which condenses 2048 associative processors, each of 128 ''intelligent'' bits. Each bit can be a processing bit or st memory bit. At only 33 Mhz and 0.6 micron manufacturing technology process, the chip has It computational power of 3 Billion ALU operations per second and 66 Billion string search operations per second. The fully programmable nature of the Xium(TM)-2 chip enables developers to use ACL tools to write their own proprietary algorithms combined with existing imageprocessing and analysis functions from ACL's extended set of libraries.
Restoration by deconvolution of three-dimensional images that have been contaminated by noise and spatially invariant blur is computationally demanding. We describe efficient parallel implementations of iterative meth...
详细信息
ISBN:
(纸本)0819450782
Restoration by deconvolution of three-dimensional images that have been contaminated by noise and spatially invariant blur is computationally demanding. We describe efficient parallel implementations of iterative methods for image deconvolution on a distributed memory computing cluster.
The communication overhead in many multiprocessor computing platform is a critical factor over performance. In this paper we will present communication performance of a large processing array built with TI 320C40 DSPs...
详细信息
ISBN:
(纸本)0819425885
The communication overhead in many multiprocessor computing platform is a critical factor over performance. In this paper we will present communication performance of a large processing array built with TI 320C40 DSPs. Inter-processor communication is provided by message passing which is a common method used in multiprocessors systems. The system is developed for imageprocessing therefore transmission of large data blocks and various forms of communication are required frequently. The processor used in this system has six built in communication links. They are X-bit, bi-directional links with a speed of 20 Mbytes/sec. A processing array built with these processors employs MIMD paradigm and static interconnection. In this paper, the communication performance of such DSP network is investigated and performance results are presented. The communication functions include broadcasting, scattering, gathering and point to point transmission of messages.
Owing to the sequential nature of memory interfaces, as well as the growing processor-memory performance gap, the design of parallelimage processors is often faced with a challenge in deciding memory organisation and...
详细信息
Owing to the sequential nature of memory interfaces, as well as the growing processor-memory performance gap, the design of parallelimage processors is often faced with a challenge in deciding memory organisation and distribution. This work addresses the problem of memory access bottlenecks in parallel digital image processors and presents one solution which demonstrates up to 93.4% reduction over standard sequential methods.
暂无评论