A high-speed stereovision system based on smart cameras is presented to track table tennis ball. A distributedparallelprocessing architecture is developed to improve the real-time performance of the system. A set of...
详细信息
Analysis of biomedical images requires attention to image features that represent a small fraction of the total image size. A rapid method for eliminating unnecessary detail, analogous to pre-attentive processing in b...
详细信息
ISBN:
(纸本)9781424432950
Analysis of biomedical images requires attention to image features that represent a small fraction of the total image size. A rapid method for eliminating unnecessary detail, analogous to pre-attentive processing in biological vision, allows computational resources to be applied where most needed for higher-level analysis. In this report we describe a method for bottom up merging of pixels into larger units based on flexible saliency criteria using a method similar to structured adaptive grid methods used for solving differential equations on physical domains. While creating a multiscale quadtree representation of the image, a saliency test is applied to prune the tree to eliminate unneeded details, resulting in an image with adaptive resolution. This method may be used as a first step for image segmentation and analysis and is inherently parallel, enabling implementation on programmable hardware or distributed memory clusters.
Microscopic imaging is an important tool for characterizing tissue morphology and pathology. 3D reconstruction and visualization of large sample tissue structure requires registration of large sets of high-resolution ...
详细信息
Microscopic imaging is an important tool for characterizing tissue morphology and pathology. 3D reconstruction and visualization of large sample tissue structure requires registration of large sets of high-resolution images. However, the scale of this problem presents a challenge for automatic registration methods. In this paper we present a novel method for efficient automatic registration using graphics processing units (GPUs) and parallel programming. Comparing a C++ CPU implementation with Compute Unified Device Architecture (CUDA) libraries and pthreads running on GPU we achieve a speed-up factor of up to 4.11x with a single GPU and 6.68x with a GPU pair. We present execution times for a benchmark composed of two sets of large-scale images: mouse placenta (16K x16K pixels) and breast cancer tumors (23K x62K pixels). It takes more than 12 hours for the genetic case in C++ to register a typical sample composed of 500 consecutive slides, which was reduced to less than 2 hours using two GPUs, in addition to a very promising scalability for extending those gains easily on a large number of GPUs in a distributed system.
Three-dimensional (3D) geophysical imaging is now receiving considerable attention for electrical conductivity mapping of potential offshore oil and gas reservoirs. The imaging technology employs controlled source ele...
详细信息
Three-dimensional (3D) geophysical imaging is now receiving considerable attention for electrical conductivity mapping of potential offshore oil and gas reservoirs. The imaging technology employs controlled source electromagnetic (CSEM) and magnetotelluric (MT) fields and treats geological media exhibiting transverse anisotropy. Moreover when combined with established seismic methods, direct imaging of reservoir fluids is possible. Because of the size of the 3D conductivity imaging problem, strategies are required exploiting computational parallelism and optimal meshing. The algorithm thus developed has been shown to scale to tens of thousands of processors. In one imaging experiment, 32,768 tasks/processors on the IBM Watson Research Blue Gene/L supercomputer were successfully utilized. Over a 24 hour period we were able to image a large scale field data set that previously required over four months of processing time on distributed clusters based on Intel or AMD processors utilizing 1024 tasks on an InfiniBand fabric. Electrical conductivity imaging using massively parallel computational resources produces results that cannot be obtained otherwise and are consistent with timeframes required for practical exploration problems.
This paper presents an overview of low level parallelimageprocessing algorithms and their implementation for active vision systems. Authors have demonstrated novel low level imageprocessing algorithms for point ope...
详细信息
This paper presents an overview of low level parallelimageprocessing algorithms and their implementation for active vision systems. Authors have demonstrated novel low level imageprocessing algorithms for point operators, local operators, dithering, smoothing, edge detection, morphological operators, image segmentation and image compression. The algorithms have been prepared & described as pseudo codes. These algorithms have been simulated using parallel Computing Toolboxtrade (PCT) of MATLAB. The PCT provides parallel constructs in the MATLAB language, such as parallel for loops, distributed arrays and message passing & enables rapid prototyping of parallel code through an interactive parallel MATLAB session.
Analysis of biomedical images requires attention to image features that represent a small fraction of the total image size. A rapid method for eliminating unnecessary detail, analogous to pre-attentive processing in b...
详细信息
Analysis of biomedical images requires attention to image features that represent a small fraction of the total image size. A rapid method for eliminating unnecessary detail, analogous to pre-attentive processing in biological vision, allows computational resources to be applied where most needed for higher-level analysis. In this report we describe a method for bottom up merging of pixels into larger units based on flexible saliency criteria using a method similar to structured adaptive grid methods used for solving differential equations on physical domains. While creating a multiscale quadtree representation of the image, a saliency test is applied to prune the tree to eliminate unneeded details, resulting in an image with adaptive resolution. This method may be used as a first step for image segmentation and analysis and is inherently parallel, enabling implementation on programmable hardware or distributed memory clusters.
The paper is directed to the advanced Quasi Monte Carlo methods for realistic image synthesis. We propose and consider a new Quasi Monte Carlo solution of the rendering equation by uniform quadrangle separation of int...
The paper is directed to the advanced Quasi Monte Carlo methods for realistic image synthesis. We propose and consider a new Quasi Monte Carlo solution of the rendering equation by uniform quadrangle separation of integration domain. The hemispherical integration domain is uniformly separated into 12 equal size and symmetric sub‐domains. Each sub‐domain represents a solid angle, subtended by spherical quadrangle, very similar by form to plane unit square. Any spherical quadrangle has fixed vertices and computable parameters. A bijection of unit square into spherical quadrangle is find and the symmetric sampling scheme is applied to generate the sampling points uniformly distributed over hemispherical integration domain. Then, we apply the stratified Quasi Monte Carlo integration method for solving the rendering equation. The estimate of the rate of convergence is obtained. We prove the superiority of the proposed Quasi Monte Carlo solution of the rendering equation for arbitrary dimension of the sampling points. The uniform separation leads to convergence improvement of the Plain (Crude) Quasi Monte Carlo method.
We develop a novel approach for computing the circle Hough transform entirely on graphics hardware (GPU). A primary role is assigned to vertex processors and the rasterizer, overshadowing the traditional foreground of...
详细信息
We develop a novel approach for computing the circle Hough transform entirely on graphics hardware (GPU). A primary role is assigned to vertex processors and the rasterizer, overshadowing the traditional foreground of pixel processors and enhancing parallelprocessing. Resources like the vertex cache or blending units are studied too, with our set of optimizations leading to extraordinary peak gain factors exceeding 358x over a typical CPU execution. Software optimizations, like the use of precomputed tables or gradient information and hardware improvements, like hyperthreading and multicores are explored on CPUs as well. Overall, the GPU exhibits better scalability and much greater parallel performance to become a solid alternative for computing the classical circle Hough transform versus those optimal methods run on emerging multicore architectures. (c) 2008 Elsevier Inc. All rights reserved.
暂无评论