Silhouettes contain rich information about the shape of objects that can be used for recognition and classification. We present a novel approach that allows us to reliably compute many useful properties of a silhouett...
详细信息
Silhouettes contain rich information about the shape of objects that can be used for recognition and classification. We present a novel approach that allows us to reliably compute many useful properties of a silhouette. Our approach assigns for every internal point of the silhouette a value reflecting the mean time required for a random walk beginning at the point to hit the boundaries. This function can be computed by solving Poisson's equation, with the silhouette contours providing boundary conditions. We show how this function can be used to reliably extract various shape properties including part structure and rough skeleton, local orientation and aspect ratio of different parts, and convex and concave sections of the boundaries. In addition to this we discuss properties of the solution and show how to efficiently compute this solution using multi-grid algorithms. We demonstrate the utility of the extracted properties by using them for shape classification.
Many lock-free data structures in the literature exploit techniques that are possible only because state-of-the-art 64-bit processors are still running 32-bit operating systems and applications. As software catches up...
详细信息
ISBN:
(纸本)9781581138023
Many lock-free data structures in the literature exploit techniques that are possible only because state-of-the-art 64-bit processors are still running 32-bit operating systems and applications. As software catches up to hardware, "64-bit-clean" lock-free data structures, which cannot use such techniques, are needed. We present several 64-bit-clean lock-free implementations: load-linked/store-conditional variables of arbitrary size, a FIFO queue, and a freelist. In addition to being portable to 64-bit software, our implementations also improve on previous ones in that they are space-adaptive and do not require knowledge of the number of threads that will access them.
In this paper, we discuss an efficient parallel implementation of the treecode Ewald method for fast evaluation of long-range Coulomb interactions in a periodic system for molecular dynamics simulations. The paralleli...
详细信息
ISBN:
(纸本)0769521320
In this paper, we discuss an efficient parallel implementation of the treecode Ewald method for fast evaluation of long-range Coulomb interactions in a periodic system for molecular dynamics simulations. The parallelization is based on an adaptive decomposition scheme using the Morton order of the particles. This decomposition scheme takes advantage of the data locality and involves minimum changes to the original sequential code. The Message Passing Interface (MPI) is used for inter-processor communications, making the code portable to a variety of parallel computing platforms. We also discuss communication and performance models for our parallel algorithm. The predicted communication time and parallel performance from these models match the measured results well. Timing results obtained using a system of water molecules on the IA32 Cluster at the Ohio Supercomputer Center demonstrate high speedup and efficiency of the parallel treecode Ewald method.
This paper provides a brief development roadmap of the neural network sensitivity analysis, from 1960's to now on. The two main streams of the sensitivity measures: partial derivative and stochastic sensitivity me...
详细信息
ISBN:
(纸本)0780384032
This paper provides a brief development roadmap of the neural network sensitivity analysis, from 1960's to now on. The two main streams of the sensitivity measures: partial derivative and stochastic sensitivity measures are compared. The partial derivative sensitivity (PD-SM) finds the rate of change of the network output with respect to parameter changes, while the stochastic sensitivity (ST-SM) finds the magnitudes of the output perturbations between the original training samples and the perturbed samples, in statistical sense. Their computational complexities are compared. Furthermore, how to evaluate multiple parameters of the neural network with or without correlation will be explored too. In addition, the differences of them in the application of supervised pattern classification problems are also discussed. The evaluations are based on three major applications of sensitivity analysis in supervised pattern classification problems: feature selection, sample selection and neural network generalization assessment ST-SM and PD-SM of the RBFNN will be used for investigations.
In order to understand AdaBoost's dynamics, especially its ability to maximize margins, we derive an associated simplified nonlinear iterated map and analyze its behavior in low-dimensional cases. We find stable c...
详细信息
ISBN:
(纸本)0262201526
In order to understand AdaBoost's dynamics, especially its ability to maximize margins, we derive an associated simplified nonlinear iterated map and analyze its behavior in low-dimensional cases. We find stable cycles for these cases, which can explicitly be used to solve for Ada- Boost's output. By considering AdaBoost as a dynamical system, we are able to prove Ratsch and Warmuth's conjecture that AdaBoost may fail to converge to a maximal-margin combined classifier when given a 'nonoptimal' weak learning algorithm. AdaBoost is known to be a coordinate descent method, but other known algorithms that explicitly aim to maximize the margin (such as AdaBoost and arc-gv) are not. We consider a differentiable function for which coordinate ascent will yield a maximum margin solution. We then make a simple approximation to derive a new boosting algorithm whose updates are slightly more aggressive than those of arcgv.
Beginning undergraduate engineering students need computing skills. A course was developed to support the retention of those undergraduates who were lacking in basic computing skills. This paper discusses the results ...
详细信息
Beginning undergraduate engineering students need computing skills. A course was developed to support the retention of those undergraduates who were lacking in basic computing skills. This paper discusses the results of using novel technologies in an assimilation course. The technologies consisted of autonomous robots that were relatively easy to build and Pocket PCs with an integrated wireless capability in order to access the Internet. The paper includes the background information on the course, technologies, and the applicability. In addition, student experiences with the tools are also covered in the paper.
In the paper the numerical analysis of thermal processes proceeding in the domain of biological tissue subjected to an external heat source is presented. Heat transfer in the skin tissue was assumed to be transient an...
详细信息
In the paper the numerical analysis of thermal processes proceeding in the domain of biological tissue subjected to an external heat source is presented. Heat transfer in the skin tissue was assumed to be transient and two-dimensional. The bioheat transfer in the domain considered is described by the system of Pennes equations determining the temperature field in successive skin layers. Between the layers the ideal contact is assumed. On the selected part of skin surface the Neumann condition determining the value of external heat source is given, on the conventionally assumed internal surface of the tissue the no-flux condition is accepted. For time t = 0 the initial distribution of temperature is known. The degree of the skin burn can be predicted on the basis of the so-called Henriques integrals and the main subject of the paper is the sensitivity analysis of these integrals with respect to the skin parameters. On the stage of numerical computations the boundary element method has been used. In the final part of the paper the results obtained are shown.
In this paper, a novel approach of rough set-based case-based reasoning (CBR) approach is proposed to tackle the task of text categorization (TC). The initial work of integrating both feature and document reduction/se...
详细信息
ISBN:
(纸本)0780384032
In this paper, a novel approach of rough set-based case-based reasoning (CBR) approach is proposed to tackle the task of text categorization (TC). The initial work of integrating both feature and document reduction/selection in TC using rough sets and CBR properties is presented. Rough set theory is incorporated to reduce the number of feature terms through generating reducts. On the other hand, two concepts of case coverage and case reachability in CBR are used in selecting representative documents. The main contribution of this paper is that both the number of features and the documents are reduced with minimal loss of useful information. Some experiments are conducted on the text datasets of Reuters21578. The experimental results show that, although the number of feature terms and documents are reduced greatly, the problem-solving quality in terms of classification accuracy is still preserved.
It is proved that the optimal solution for the matching problem in multi-target tracking can be found from among n different matching. We show in this paper by simulation that the costs of the n different possible sol...
详细信息
ISBN:
(纸本)0780384032
It is proved that the optimal solution for the matching problem in multi-target tracking can be found from among n different matching. We show in this paper by simulation that the costs of the n different possible solutions constitute a bimodal sequence, which suggests an algorithm of O(N logN) complexity, lower than most of known algorithms.
暂无评论