The generalized matrix product includes in its formulation many common array manipulations. It also provides a framework for the expression of a number of important imageprocessing algorithms. It is shown that the ge...
详细信息
In this paper, we propose a framework for the efficient implementation of two-dimensional (2-D) noncausal infinite impulse response iiR) filters, i.e,, filter systems described implicitly by difference equations and b...
详细信息
In this paper, we propose a framework for the efficient implementation of two-dimensional (2-D) noncausal infinite impulse response iiR) filters, i.e,, filter systems described implicitly by difference equations and boundary conditions, A number of common 2-D LSI filter operations, (including lowpass, high-pass, and zero-phase filters), are efficiently realized and implemented in this paper as noncausal iiR filters, The basic concepts involved in our approach include the adaptation of so-called direct methods for solving partial differential equations (PDE's), and the introduction of an approximation methodology that is particularly well suited to signal processing applications and leads to very efficient implementations. In particular, for an input and output with N x N samples, the algorithm requires only O(N-2) storage and computations (yielding a per pixel computational load that is independent of image size), and has a parallel implementation (yielding a per pixel computational load that decreases with increasing image size), Also, because our approach allows for the implementation of filters with space-varying coefficients on irregularly shaped domains, it should have applications in related areas like linear estimation, geophysical signal processing, or any held requiring approximate solutions to elliptic PDE's.
Edge detection is an important first step in many vision tasks where its improvements in speed and efficiency present a continuous challenge for developers of high-speed image recognizers. Classical techniques for acc...
详细信息
Given a Cartesian product G = G(l) x ... x G(m)(m greater than or equal to 2) of nontrivial connected graphs Gi and the n-dimensional base B de Bruijn graph D = D-B(n), it is investigated whether or not G is a spannin...
详细信息
Given a Cartesian product G = G(l) x ... x G(m)(m greater than or equal to 2) of nontrivial connected graphs Gi and the n-dimensional base B de Bruijn graph D = D-B(n), it is investigated whether or not G is a spanning subgraph of D. Special attention is given to graphs G(l) x ... x G(m) which are relevant for parallel computing, namely, to Cartesian products of paths (grids) or cycles (tori), For 2-dimensional de Bruijn graphs D, we present a theorem stating that certain structural assumptions on the factors G(l), ..., G(m) ensure that G(l) x ... x G(m) is a spanning subgraph of D. As corollaries, we obtain results improving previous results of Heydemann et al. (J. parallel Distrib. Compat. 23 (1994), 104-111) on embedding grids and tori into de Bruijn graphs. Specifically, we obtain that (i) any grid G = G(l) x ... x G(m) is a spanning subgraph of D = D-B(2) provided that \G\ = \D\, and (ii) any torus G = G(l) x ... x G(m) is a spanning subgraph of D = D-B(2) provided that \G\ = \D\ and that the G(l) are cycles of even length greater than or equal to 4. We show that these results have consequences for the case a >2, too: for even n, we apply our results to obtain embeddings of grids and tori G into de Bruijn graphs D-B(n) with dilation n/2, where the base B is a fixed integer greater than or equal to 2, and n is big enough to ensure \G\ less than or equal to \D-B(n)\. We also contrast our results for n = 2 with nonexistence results for n greater than or equal to 3 and briefly describe experimental results in the area of parallelimageprocessing. (C) 1997 Academic Press.
We describe a dynamic load-balancing algorithm for ray-tracing by progressive refinement on a distributed-memory parallel computer. parallelization of progressive ray-tracing for single images is difficult because of ...
详细信息
ISBN:
(纸本)1581130104
We describe a dynamic load-balancing algorithm for ray-tracing by progressive refinement on a distributed-memory parallel computer. parallelization of progressive ray-tracing for single images is difficult because of the inherent sequential nature of the sample location generation process, which is optimized (and different) for any given image. parallelization of progressive ray-tracing when generating image sequences at a fixed interactive rate is even more difficult, because of the time and synchronization constraints imposed on the system. We show how to overcome these problems, which, to the best of our knowledge, have not been treated before. Exploiting the temporal coherence between frames enables us to both accelerate rendering and improve the load-balance throughout the sequence. Our dynamic load-balance algorithm, a blend of local and global methods, accounting not only for rendering performance, but also communication overheads and synchronization issues, is shown to be robust to the harsh environment imposed by a time-critical application, such as the one we consider.
Tomographycal methods of image reconstruction of two-dimensional objects, surfaces or subsurface regions in millimeter wavelengths band are suggested and considered. Experimental images obtained using antennas and wav...
详细信息
Tomographycal methods of image reconstruction of two-dimensional objects, surfaces or subsurface regions in millimeter wavelengths band are suggested and considered. Experimental images obtained using antennas and waveguiding lines of different types and radiation frequency f approximate to 33 divided by 35GHz are represented. Volumetric dielectric objects and plane-parallel ferrite (or dielectric) plates distributed in free space or in homogeneous dielectric medium have been taken as objects under investigation. It is shown that in the frequency band under consideration, the images of investigated objects with characteristic dimension A approximate to 2 lambda divided by lambda may be obtained by first-order diffraction tomography method (Born, Rytov or high frequency approximation of the first-order for electromagnetic field).
A new algorithm to compute the motion vector has been developed based on the block-matching approach in which bands are used instead of blocks. The new band-matching algorithm (BDMA) offers better performance than bot...
详细信息
This paper describes the design concept and the implementation of the imaging database Online image Delivery System (OLIDS). We propose a new framework to support vertically parallelimageprocessing in multimedia dat...
详细信息
ISBN:
(纸本)0818681489
This paper describes the design concept and the implementation of the imaging database Online image Delivery System (OLIDS). We propose a new framework to support vertically parallelimageprocessing in multimedia database systems. We present a generic and customizable framework for specifying image query execution plan that permits the DBMS to optimize and to execute parallelimageprocessing. Moreover, it is important to provide to the multimedia applications the capability to customize the information server according to their specific requirements. The capability to customize easily in a vertical way the multimedia information server according to new applications' requirements (i.e. new kinds of retrieval methods with uncertainty factor, new kinds of compression algorithms) also provides both challenges and opportunities in building high performance multimedia and image systems. Applications such as (Active hypermedia project) or MODOS (Museum project between LRMF (France), NMWA and NACSIS(Japan)) demonstrate the usefulness of the customizable feature of the Application-Oriented DBMS Phasme, a micro-kernel database system tested and extended inside NACSIS.
Watershed transformation is a tool for image segmentation widely used in computer vision applications. However, the complexity of processing large images entails fast parallel algorithms. In this paper, an improved SP...
详细信息
The objective of thinning is to reduce the amount of information in image patterns to the minimum needed for recognition. Thinned image helps the extraction of important features such as end points, junction points, a...
详细信息
暂无评论