Fast image encryption algorithm based on parallel permutation-and-diffusion strategy

Multimedia Tools and Applications - In the image encryption algorithms, expansion of size and increase of data lead to a large amount of time consumption. To improve the efficiency, the current...     

A global optimization strategy based on the Kriging surrogate model and parallel computing

Structural and Multidisciplinary Optimization - The computational efficiency and accuracy of the global solution are the main performance indicators of an optimization algorithm to solve the...     

针对复杂系统并行模拟问题的并发式多级矩阵重排算法

在解决复杂化工过程优化与模拟问题时,大规模代数差分方程的存在导致大量的计算时间花费在重复求解稀疏大型线性方程组的过程中。随着并行计算和网络技术的发展,为了提高优化或模拟工作的速度,可以通过将非对称矩阵重排为带边块对角形式,从而实现对线性系统的高效并行求解。本文提出一种基于Kernighan-Lin算法的并发式的多层次矩阵重排策略,它以最小化边块为目标,同时保证尽可能小的负荷非平衡性,从而获得好的重排结果。应用该重排策略可以对大型稀疏矩阵进行压缩和并行重排,提高重排算法的效率。在研究过程中应用了基于该技术的并行计算程序对一系列标准矩阵进行了检验,并与一些现有的算法进行了比较,证明了其有效性和可行性。     

Learning parallel portfolios of algorithms

A wide range of combinatorial optimization algorithms have been developed for complex reasoning tasks. Frequently, no single algorithm outperforms all the others. This has raised interest in leveraging the performance of a collection of algorithms to improve performance. We show how to accomplish this using a Parallel Portfolio of Algorithms (PPA). A PPA is a collection of diverse algorithms for solving a single problem, all running concurrently on a single processor until a solution is produced. The performance of the portfolio may be controlled by assigning different shares of processor time to each algorithm. We present an effective method for finding a PPA in which the share of processor time allocated to each algorithm is fixed. Finding the optimal static schedule is shown to be an NP-complete problem for a general class of utility functions. We present bounds on the performance of the PPA over random instances and evaluate the performance empirically on a collection of 23 state-of-the-art SAT algorithms. The results show significant performance gains over the fastest individual algorithm in the collection.      

Feature fusion: parallel strategy vs. serial strategy

A new strategy of parallel feature fusion is introduced in this paper. A complex vector is first used to represent the parallel combined features. Then, the traditional linear projection analysis methods, including principal component analysis, K-L expansion and linear discriminant analysis, are generalized for feature extraction in the complex feature space. Finally, the developed parallel feature fusion methods are tested on CENPARMI handwritten numeral database, NUST603 handwritten Chinese character database and ORL face image database. The experimental results indicate that the classification accuracy is increased significantly under parallel feature fusion and also demonstrate that the developed parallel fusion is more effective than the classical serial feature fusion.     

Design concepts for an on-board parallel image processor

A compact parallel image processing system concept has been developed. The main features of this system is the use of off-axis paraboloidal mirror segments as collimating, Fourier transforming and image reconstructing elements, and the use of a GaAs laser diode as the coherent radiation source. Preliminary experiments to demonstrate the usefulness of this system have been performed.     

Task division for parallel implementation of object identification system based on alternating hypothesize-verify-extend strategy

Parallelism has been perceived as the way for a computer vision system to achieve the required speedup in practice with existing algorithms and computing resources. It is known that methods of task division in such systems are algorithm-dependent. This paper investigates the parallel implementation of an object identification system based on an alternating hypothesize-verify-extend strategy with a transputer mesh network. This non-deterministic task is firstly divided on the hypothesis level and then within the processing for each hypothesis. It has been demonstrated that an object with modest complexity can be recognized and located from a cluttering image in a fraction of a second. © 1997 John Wiley & Sons, Ltd.     

Towards the development of an h–p-refinement strategy based upon error estimate sensitivity

The use of (a posteriori) error estimates is a fundamental tool in the application of adaptive numerical methods across a range of fluid flow problems. Such estimates are incomplete however, in that they do not necessarily indicate where to refine in order to achieve the most impact on the error, nor what type of refinement (for example h-refinement or p-refinement) will be best. This paper extends preliminary work of the authors (Comm Comp Phys, 2010;7:631–8), which uses adjoint-based sensitivity estimates in order to address these questions, to include application with p-refinement to arbitrary order and the use of practical a posteriori estimates. Results are presented which demonstrate that the proposed approach can guide both the h-refinement and the p-refinement processes, to yield improvements in the adaptive strategy compared to the use of more orthodox criteria.     

Learning spatial concepts from RatSLAM representations

RatSLAM is a biologically-inspired visual SLAM and navigation system that has been shown to be effective indoors and outdoors on real robots. The spatial representation at the core of RatSLAM, the experience map, forms in a distributed fashion as the robot learns the environment. The activity in RatSLAM’s experience map possesses some geometric properties, but still does not represent the world in a human readable form. A new system, dubbed RatChat, has been introduced to enable meaningful communication with the robot. The intention is to use the “language games” paradigm to build spatial concepts that can be used as the basis for communication. This paper describes the first step in the language game experiments, showing the potential for meaningful categorization of the spatial representations in RatSLAM.     

Learning and classification of monotonic ordinal concepts

Ordinal reasoning plays a major role in human cognition. This paper identifies an important class of classification problems of patterns taken from ordinal domains and presents efficient, incremental algorithms for learning the classification rules from examples. We show that by adopting a monotonicity assumption of the output with respect to the input, inconsistencies among examples can be easily detected and the number of possible classification rules substantially reduced. By adopting a conservative classification criterion, the required number of rules further decreases. The monotonicity and conservatism of the classification also enable the resolution of conflicts among inconsistent examples and the graceful handling of don't knows and don't cares during the learning and classification phases. Two typical examples in which the suggested classification model works well are given. The first example is taken from the financial domain and the second from machining.     

云中面向图像并行计算的数据分布策略

有效地减少云计算系统中对计算任务的处理响应时间,并使各计算机节点负载均衡,数据分布算法是相当重要的.提出了一种面向图像并行计算的适用于主从类型云计算系统结构的数据分布策略,设计节点性能函数来表示节点的处理能力,根据节点间的性能比率进行任务数据量的分布,结合链路带宽制定数据发送的顺序.模拟实验结果表明,该算法适用于云计算环境,能明显提高系统的数据处理效率.     

基于改进DE算法的并联泵组能效最优控制策略研究

为改变传统给水系统在节能降耗方面的不足,结合供水系统的典型特点,建立了以节能为目标的水泵并联运行优化调度模型;采用差分进化(DE)算法对模型进行求解,在确保供水系统服务用户质量的前提下,优化调度泵站开启水泵的台数和调速率,使系统整体运行的能耗最低.使用MATLAB搭建了仿真实验平台,验证了差分进化算法的有效性和实用性.     

Learning distributed representations of concepts using linearrelational embedding

We introduce linear relational embedding as a means of learning a distributed representation of concepts from data consisting of binary relations between these concepts. The key idea is to represent concepts as vectors, binary relations as matrices, and the operation of applying a relation to a concept as a matrix-vector multiplication that produces an approximation to the related concept. A representation for concepts and relations is learned by maximizing an appropriate discriminative goodness function using gradient ascent. On a task involving family relationships, learning is fast and leads to good generalization     

A hybrid strategy to solve the forward kinematics problem in parallel manipulators

A parallel manipulator is a closed kinematic structure with the necessary rigidity to provide a high payload to self-weight ratio suitable for many applications in manufacturing, flight simulation systems, and medical robotics. Because of its closed structure, the kinematic control of such a mechanism is difficult. The inverse kinematics problem for such manipulators has a mathematical solution; however, the forward kinematics problem (FKP) is mathematically intractable. This work addresses the FKP and proposes a neural-network-based hybrid strategy that solves the problem to a desired level of accuracy, and can achieve the solution in real time. Two neural-network (NN) concepts using a modified form of multilayered perceptrons with backpropagation learning were implemented. The better performing concept was then combined with a standard Newton-Raphson numerical technique to yield a hybrid solution strategy. Simulation studies were carried out on a flight simulation syystem to check the validity o the approach. Accuracy of close to 0.01 mm and 0.01/spl deg/ in the position and orientation parameters was achieved in less than two iterations and 0.02 s of execution time for the proposed strategy.     

An efficient parallel strategy for high-cost prefix operation

The Journal of Supercomputing - The prefix computation strategy is a fundamental technique used to solve many problems in computer science such as sorting, clustering, and computer vision. A large...     

Using machine learning in a cooperative hybrid parallel strategy of metaheuristics

This paper proposes the construction of a centralized hybrid metaheuristic cooperative strategy to solve optimization problems. Knowledge (intelligence) is incorporated into the coordinator to improve performance. This knowledge is incorporated through a set of rules and models obtained from a knowledge extraction process applied to the records of the results returned by individual metaheuristics. The effectiveness of the approach is tested in several computational experiments in which we compare the results obtained by the individual metaheuristics, by several non-cooperative and cooperative strategies and by the strategy proposed in this paper.