In this study, we introduce a novel local image descriptor, which is very efficient to compute densely. We also present an algorithm to compute dense depth maps from image pairs using designed descriptor. Novel descri...
详细信息
ISBN:
(数字)9781510618527
ISBN:
(纸本)9781510618527
In this study, we introduce a novel local image descriptor, which is very efficient to compute densely. We also present an algorithm to compute dense depth maps from image pairs using designed descriptor. Novel descriptor is based on visual primitives and relations between them, namely coplanarity, cocolority, distance, and angle. Designed feature descriptor covers both geometric and appearance information. The depth map estimation performance is evaluated using the established bad matched pixel metric. An analysis of the feature descriptor employing a parallel programming paradigm is included to develop a possible real-time mode. This is performed with help of hardware based on multi -core processors and GPU platform, using a NVIDIA GeForce GT640 graphic card and Matlab over a PC with Windows 10.
Pseudo-sequence based light field image compression methods use state-of-the-art video codecs like HEVC. Although video codecs have been designed to compress video sequences, they have good performance for light field...
详细信息
ISBN:
(数字)9781510620766
ISBN:
(纸本)9781510620766
Pseudo-sequence based light field image compression methods use state-of-the-art video codecs like HEVC. Although video codecs have been designed to compress video sequences, they have good performance for light field images compression. Considering the light field images represented by their multi-views representation, a sequence with different image views is aligned following an appropriate strategy. However, there are some differences between video sequences and light field pseudo videos that can be utilized to improve the codec adaptation to the different view images compression. The pseudo-sequence images have spatial distances and predictable behaviors when compared to video sequences that have temporal distance and unpredictable behavior respectively. Considering these differences unnecessary operations can be avoided, while its performance can be improved. The video codecs compute the motion vectors using block matching motion estimation algorithms, which is computationally the most complex operation of any video codec. To reduce the motion estimation complexity many codecs use prediction models. In this paper, HEVC motion vectors search models are applied to the light field image views aligned as pseudo-sequences to analyze and find their repetitive and predictable patterns. These new patterns are then utilized for changing the HEVC motion estimation algorithm for codec complexity reduction using a state-of-the-art pseudo sequence compression method. Moreover, the use of parallel computing for the pseudo sequence compression method is addressed.
Heterogeneous information networks (e.g. cloud service relation networks and social networks), where multiple-typed objects are interconnected, can be structured by big graphs. A major challenge for clustering in such...
详细信息
Heterogeneous information networks (e.g. cloud service relation networks and social networks), where multiple-typed objects are interconnected, can be structured by big graphs. A major challenge for clustering in such big graphs is the complex structures that can generate different results, carrying many diverse semantic meanings. In order to generate desired clustering, we propose a parallel clustering method for the heterogeneous information net-works on an efficient graph computation system (Spark). We use a multi-relation and path-based method to create similarity matrices, and implement our method based on graph computation model. It is inefficient to directly use existing data-parallel tools (e.g. Hadoop) for graph computation tasks, and some graph-parallel tools (e.g. Pregel) do not effectively address the challenges of graph construction and transformation. Therefore, we implemented our parallel method on the Spark system. The experiment results of clustering show our method is more accuracy.
A source-load coordination scheduling strategy is proposed in this paper to reduce the system operation cost and wind power curtailment. Firstly, the scheduling model of the power system with wind power is established...
详细信息
ISBN:
(纸本)9781862203426
A source-load coordination scheduling strategy is proposed in this paper to reduce the system operation cost and wind power curtailment. Firstly, the scheduling model of the power system with wind power is established. To solve the scheduling problem, the binary particle swarm optimization (BPSO) algorithm is used to determine the ON/OFF states of generations;the continuous particle swarm optimization (CPSO) algorithm is used to deal with the economic load dispatch problem;and the constraints are properly handled by adjustment methods. Secondly, in order to maximize the wind power accommodation rate, the power system adopts the time-of-use price program, an optimization model of electricity price is established based on price elasticity matrix. The CPSO algorithm and parallel computing are used to optimize the time-of-use price schedules. According to the results of the case study, the demand response program plays an important role in reducing the peak-valley difference, wind power curtailment, and system operating cost. The proposed scheduling strategy and algorithm are proven to have a good optimization performance, calculation speed and stability.
Investigating the improvement of MATLAB based finite difference time domain (FDTD) simulations using central processing units (CPUs) is the goal of this paper. The MATLAB parallel toolbox along with modern multi-core ...
详细信息
ISBN:
(纸本)9781728115184;9780996007887
Investigating the improvement of MATLAB based finite difference time domain (FDTD) simulations using central processing units (CPUs) is the goal of this paper. The MATLAB parallel toolbox along with modern multi-core CPUs are used with no external dependencies or libraries and no expensive or complicated hardware acceleration units.
In view of the high reliability and high efficiency of parallel computing system for parallel computing, according to the basic methods and ideas of computer system reliability and fault-tolerant technology, the desig...
详细信息
ISBN:
(纸本)9781538653739
In view of the high reliability and high efficiency of parallel computing system for parallel computing, according to the basic methods and ideas of computer system reliability and fault-tolerant technology, the design of multi thread and multi thread parallel fault-tolerant computer system is proposed and the midway message and isolation in the calculation process are presented. The corresponding model design and technical evaluation work are carried out. The simulation verification shows that the design of multi thread parallel computing fault tolerance model can effectively improve the reliability of the system.
In this paper, a Hardware-in-the-loop (HIL) simulation method for multiple fuel cells hybrid power system based on field-programmable gate array (FPGA) + CPU architecture is presented. This paper is aiming to propose ...
详细信息
In this paper, a Hardware-in-the-loop (HIL) simulation method for multiple fuel cells hybrid power system based on field-programmable gate array (FPGA) + CPU architecture is presented. This paper is aiming to propose a parallel modular computing strategy in FPGA, which is used to reduce the programming difficulty and single-step simulation time. In this strategy, different modules are modeled inside FPGA separately and are connected together to achieve the desired circuit by an interactive connection module. In this way, complex programming is not required when building a system. It only needs a combination of different modules, just like building blocks or modeling in Simulink. This strategy can reduce the programming difficulty greatly when developing complex models in FPGA. Furthermore, it is highly parallelizable to achieve a single-step simulation time of 0.1 mu s in FPGA. The simulation and experimental results demonstrate that the HIL simulation platform has the advantages of high speed, high precision, low cost, and easy to popularize.
In complex phenotypes (e.g., psychiatric diseases) single locus tests, commonly performed with genome-wide association studies, have proven to be limited in discovering strong gene associations. A growing body of evid...
详细信息
ISBN:
(纸本)9781538649756
In complex phenotypes (e.g., psychiatric diseases) single locus tests, commonly performed with genome-wide association studies, have proven to be limited in discovering strong gene associations. A growing body of evidence suggests that epistatic non-linear effects may be responsible for complex phenotypes arising from the interaction of different biological factors. A major issue in epistasis analysis is the computational burden due to the huge number of statistical tests to be performed when considering all the potential genotype combinations. In this work, we developed a computational efficient approach to compute empirical p-values concerning the presence of epistasis at a genome-wide scale in bipolar disorder, which is a typical example of complex phenotype with a relevant but unexplained genetic background. By running our approach we were able to identify 13 epistasis interactions between variants located in genes potentially involved in biological processes associated with the analyzed phenotype.
With the requirement that parallel & distributed computing (PDC) topics be covered in the core computer science curriculum, educators are exploring new ways to engage students in this area of computing. In this pa...
详细信息
ISBN:
(纸本)9781450351034
With the requirement that parallel & distributed computing (PDC) topics be covered in the core computer science curriculum, educators are exploring new ways to engage students in this area of computing. In this paper, we discuss the use of the Raspberry Pi single-board computer (SBC) to provide students with hands-on multicore learning experiences. We discuss how the authors use the Raspberry Pi to teach parallel computing, and present assessment results that indicate such devices are effective at achieving CS2013 PDC learning outcomes, as well as motivating further study of parallelism. We believe our results are of significant interest to CS educators looking to integrate parallelism in their classrooms, and support the use of other SBCs for teaching parallel computing.
Attack graph is used as a model that enumerates all possible attack paths based on a comprehensive analysis of multiple network configurations and vulnerability information. An attack graph generation method based on ...
详细信息
ISBN:
(纸本)9783030030261;9783030030254
Attack graph is used as a model that enumerates all possible attack paths based on a comprehensive analysis of multiple network configurations and vulnerability information. An attack graph generation method based on parallel computing is therefore proposed to solve the thorny problem of calculations as the network scale continues to expand. We utilize multilevel k-way partition algorithm to divide network topology into parts in efficiency of parallel computing and introduce Spark into the attack graph generation as a parallel computing platform. After the generation, we have a tool named Monitor to regenerate the attack graph of the changed target network. The method can improve the speed of calculations to solve large and complex computational problems and save time of generating the whole attack graph when the network changed. The experiments which had been done show that the algorithm proposed to this paper is more efficient benefiting from smaller communication overhead and better load balance.
暂无评论