A compact and highly programmable processor based on the architecture of the ICL Distributed Array Processor (DAP) is under current development for real time signal processing.applications. The architecture is describ...
详细信息
A compact and highly programmable processor based on the architecture of the ICL Distributed Array Processor (DAP) is under current development for real time signal processing.applications. The architecture is described and the applicability of this highly concurrent machine is demonstrated with specific reference to algorithms in speech recognition, radar, signal classification and imageprocessing. The results indicate that the performance compares well with existing dedicated digital signal processing.hardware.
In this paper, a comprehensive overview of Computer architecture for digital Signal processing.is given. Such architectures are seen as the result of constraining influences from the nature of digital signal processin...
详细信息
In this paper, a comprehensive overview of Computer architecture for digital Signal processing.is given. Such architectures are seen as the result of constraining influences from the nature of digital signal processing.algorithms, architectural techniques including appropriate choice of primitive elements, the underlying digital system technology, and programming languages for digital signal processing. Following a consideration of these influences, several examples are given ranging from chips through board level processors, to attached support processors with very high throughput. Trends for the future are discussed throughout the paper.
The system illustrated in this paper has been designed and developed particularly for automatic and reliable analysis of body movement in various conditions and environments. It is based on real-time processing.of the...
详细信息
The system illustrated in this paper has been designed and developed particularly for automatic and reliable analysis of body movement in various conditions and environments. It is based on real-time processing.of the TV images to recognize multiple passive markers and compute their coordinates. This performance is achieved by using a special algorithm allowing the recognition of markers only if their shape matches a predetermined "mask." The main feature of the system is a two-level processing.architecture, the first of which includes a dedicated peripheral fast processor for shape recognition (FPSR), designed and implemented by using fast VLSI chips. The second level consists of a general purpose computer and provides the overall system with high flexibility. The main characteristics are: no restriction on the number of markers, resolution of one part in 2500, and a 50 Hz sampling rate independent of the number of markers detected. The prototype has been fully developed, and preliminary results obtained from the analysis of several movements are illustrated.
This paper deals with the problem of computing, projections of digitalimages. The novelty of our contribution is that we present algorithms which are suitable for implementation in general purpose imageprocessing.an...
详细信息
This paper deals with the problem of computing, projections of digitalimages. The novelty of our contribution is that we present algorithms which are suitable for implementation in general purpose imageprocessing.and image analysis pipeline architectures. Also, we propose some new pipeline configurations which achieve a remarkable degree of parallelism in the computation of projection data and in fact, of many other geometrical descriptors of digitalimages. In particular, random access memories and other dedicated hardware devices are not needed in our algorithms. The effectiveness of our approach and feasibility of the proposed architectures are demonstrated by running our algorithms in commercially available short-pipelines for imageprocessing.and analysis. Examples are shown of the use of projection data for machine vision applications.
An increasing awareness of the need for high speed parallel processing.systems for image analysis has stimulated a great deal of interest in the design and development of such systems. Efficient processing.schemes for...
详细信息
An increasing awareness of the need for high speed parallel processing.systems for image analysis has stimulated a great deal of interest in the design and development of such systems. Efficient processing.schemes for several specific problems have been developed providing some insight into the general problems encountered in designing efficient imageprocessing.algorithms for parallel architectures. However it is still not clear what architecture or architectures are best suited for imageprocessing.in general, or how one may go about determining those which are. An approach that would allow application requirements to specify architectural features would be useful in this context. Working towards this goal, general principles are outlined for formulating parallel imageprocessing.tasks by exploiting parallelism in the algorithms and data structures employed. A synchronous parallel processing.model is proposed which governs the communication and interaction between these tasks. This model presents a uniform framework for comparing and contrasting different formulation strategies. In addition, techniques are developed for analyzing instances of this model to determine a high level specification of a parallel architecture that best ‘matches’ the requirements of the corresponding application. It is also possible to derive initial estimates of the component capabilities that are required to achieve predefined performance levels. Such analysis tools are useful both in the design stage, in the selection of a specific parallel architecture, or in efficiently utilizing an existing one. In addition, the architecture independent specification of application requirements makes it a useful tool for benchmarking applications.
The Environmental Research Institute of Michigan (ERIM) has developed the fourth generation of its cellular imageprocessing.systems, known as Cytocomputers®. These systems have been developed over the past nine ...
详细信息
The colloquium contains nine individual papers on digital signal processing.technique, which until recently has been too expensive and arithmetically intensive to achieve wide penetration in real time applications. Tw...
详细信息
The colloquium contains nine individual papers on digital signal processing.technique, which until recently has been too expensive and arithmetically intensive to achieve wide penetration in real time applications. Two dimensional least mean square algorithms are presented and applied for noise smoothing in images;a hardware design for a FFT algorithm is given to perform decimation in a frequency domain using a pipelined structure. Vector quantization as a method for image data compression is described;algorithms for stereo vision system are presented;and a variety of vocoden algorithms are outlined. Several digital techniques are examined applicable to the field of audio electronics.
The processing.of image data presents a severe challenge for two main reasons: the data sets are very large (of the order of a quarter of a megabyte in a single television frame) and effective algorithms for extractin...
详细信息
ISBN:
(纸本)0869804731
The processing.of image data presents a severe challenge for two main reasons: the data sets are very large (of the order of a quarter of a megabyte in a single television frame) and effective algorithms for extracting the data are difficult to discover or devise. It is also true that the types of operations which seem to be necessary vary enormously between low-level processing.(for image enhancement and restoration, filtering and so on) and high-level processing.(image understanding). Because of this, it might be thought that the task of designing a computer architecture suited to imageprocessing. and therefore capable of achieving high processing.rates, would be impossible. A general survey of neighbourhood operators was presented by the author at an international symposium organized by the Rank Prize Funds and forms the back-ground for the present paper.
The author observes that divide-and-conquer is an important algorithm strategy, although it is not widely used in imageprocessing. For higher-level symbolic operations, he contends that it should often be the strateg...
详细信息
ISBN:
(纸本)0818606622
The author observes that divide-and-conquer is an important algorithm strategy, although it is not widely used in imageprocessing. For higher-level symbolic operations, he contends that it should often be the strategy of choice for parallel computers. It is natural for a machine with a regular interconnection scheme such as a mesh, pyramid, or hypercube, and can be used either on a machine with one pixel per processor or one with many pixels per processor. However, divide-and-conquer algorithms use parallel computers in a different manner than, say, local edge detection, so machines optimized for local neighborhood algorithms may be poor for divide-and-conquer algorithms. Some characteristics of divide-and-conquer algorithms are examined, along with some of their implications for machine design.
暂无评论