实现虚拟现实大场景在网上的运行

利用VRML建立大范围、多景物的场景,通过Web浏览时,存在着网络带宽和绘制困难的问题。影响了虚拟场景在网上的运行效果。针对这一问题,我们必须将大范围场景按VRML的规则进行有效分割、分层调度以及性能优化。     

一种基于场景图分割的混合式多视图三维重建方法

针对大范围三维重建, 重建效率较低和重建稳定性、精度差等问题, 提出了一种基于场景图分割的大范围混合式多视图三维重建方法.该方法首先使用多层次加权核K均值算法进行场景图分割; 然后,分别对每个子场景图进行混合式重建, 生成对应的子模型, 通过场景图分割、混合式重建和局部优化等方法提高重建效率、降低计算资源消耗, 并综合采用强化的最佳影像选择标准、稳健的三角测量方法和迭代优化等策略, 提高重建精度和稳健性; 最后, 对所有子模型进行合并, 完成大范围三维重建.分别使用互联网收集数据和无人机航拍数据进行了验证, 并与1DSFM、HSFM算法在计算精度和计算效率等方面进行了比较.实验结果表明, 本文算法大大提高了计算效率、计算精度, 能充分保证重建模型的完整性, 并具备单机大范围场景三维重建能力.     

基于多层BP神经网络和无参数微调的人群计数方法

针对大部分现有的人群计数方法被应用到新的场景时性能下降的问题,在多层BP神经网络框架下,提出一种具有无参数微调的人群计数方法。首先,从训练图像中裁切图像块,将获得的相似尺度的行人作为人群BP神经网络模型的输入;然后,BP神经网络模型通过学习预测密度图,得到了一个具有代表性的人群块;最后,为了处理新场景,对训练好的BP神经网络模型进行目标场景微调,可追求有相同属性的样本,包括候选块检索和局部块检索。实验数据集包括PETS2009数据集、UCSD数据集和UCF_CC_50数据集。这些场景的实验结果验证了提出方法的有效性。相比于全局回归计数法和密度估计计数法,提出的方法在平均绝对误差和均方误差方面均有较大优势, 消除了场景间区别和前景分割的影响。     

基于场景局部特征的多曝光图像融合

针对现有多曝光图像融合算法得到的图像质量不高以及算法效率低的问题,提出了基于场景局部特征的多曝光图像融合算法。首先,将不同曝光量的图像序列划分为规则的图像块,并且相邻的图像块有一定像素的重叠区域。对于静态场景,根据图像的局部方差、局部可视性以及局部显著性特征这三个指标计算每一个图像块的权重值;对于动态场景,除了应用前面所述的三个局部特征指标外,还需要将局部相似性指标用于动态场景融合过程中以去除运动物体导致的鬼影现象。其次,利用加权求和的方法得到最佳的图像块。最后,将输出的图像块进行融合,并且将图像块重叠区域的像素求平均,从而得到最终的融合结果。选取12组不同自然场景的曝光序列,从主观和客观两方面与现有的基于像素和基于特征的7种算法进行了分析和比较。实验结果表明：无论在静态场景还是动态场景的测试中,所提算法都保留了更多的场景信息,获得了令人满意的视觉效果,同时该算法还保持了较高的计算效率。     

基于多尺度条件生成对抗网络的图像去模糊

为解决复杂场景中的非均匀模糊问题,提出一种端到端的多尺度条件生成对抗网络用于图像去模糊.在生成器网络中,针对现有的基于分层级联方式的多尺度策略会导致网络规模过大的不足,提出一种多尺度残差模块,通过在单个块内构造分级连接和增加不同感受野分支的方式来提高多尺度特征提取能力;在多尺度残差块和瓶颈层之间添加跳跃连接,自适应地提...     

一种新的人眼视觉运动认知特性实验分析方法

视觉感知过程源于对象自身的大范围拓扑性质,而场景对象发生运动状态变化时,其所在空间的大范围拓扑关系也将随之改变。为探索大范围拓扑性质和运动性质同时改变时人眼视觉的认知优先特性,基于虚拟现实技术及眼动跟踪技术,通过构建虚拟驾驶及眼动轨迹分析实验平台,结合拓扑复杂度和空间拓扑关系概念分别探讨二维和三维空间中人眼视知觉对物体运动的认知特性。在此基础上,采用响应面分析方法设计场景中自变量因子的随机实验数值,通过可视化方法对比多组认知场景的无感反馈眼动监测结果。实验结果表明,在Tobii眼动仪采集的545个眼动数据中,发生在运动物体上的数据有516个,相比于物体自身属性的大范围拓扑性质变化,人眼视知觉过程更优先识别物体运动的大范围空间拓扑关系变化。该运动认知优先特性为可靠人工智能驾驶中的视觉注视预测技术提供了新的研究思路,同时验证了所提实验分析方法的有效性。     

基于一致性敏感哈希块匹配的HDR图像去伪影融合方法

高动态范围(High dynamic range, HDR)图像成像技术的出现, 为解决由于采集设备动态范围不足而导致现有数字图像动态范围有限的问题提供了一条切实可行的思路.合成高动态范围图像的过程中因相机抖动或运动物体所造成的模糊和伪影问题, 可通过块匹配对多曝光图像序列进行去伪影融合加以解决.但对于具有复杂运动变化的真实场景, 现有的去伪影融合方法准确度和效率仍存在不足.为此, 本文结合相机响应函数和一致性敏感哈希提出了一种高动态图像去伪影融合方法.仿真结果表明, 该方法有效降低了计算复杂度, 具有较好的鲁棒性, 在有效去除伪影的同时提升了高动态范围图像质量.     

基于有向无环图的区块链技术综述

区块链技术已广泛应用于金融、公共服务、物联网、网络安全、供应链等多个领域,但传统单链结构的区块链在吞吐量、交易确认速度和可扩展性方面存在不足,导致其在一些短时高并发量数据场景中难以落地应用。基于有向无环图（DAG）的区块链技术因其具有交易并发确认功能、吞吐量高、可扩展性强等优势受到研究者广泛关注。通过分析研究现有DAG区块链的发展和演化过程、评估方法、优化方向以及应用场景,探索DAG区块链在落地应用中的可行性。阐述主流DAG区块链的研究现状,比较传统区块链与DAG区块链的优势与不足,并分析现有区块链属性评估方法,总结DAG区块链评估结果。在此基础上,从交易确认速度、系统吞吐量、系统安全性、存储结构等方面对比现有DAG区块链的优化方法,介绍DAG区块链在数据管理、基于边缘计算和联邦学习的数据共享以及面向访问控制和隐私保护的数据安全等场景中的应用情况,并指出当前研究中存在的主要问题和挑战,对未来的研究方向进行展望。     

高密度LoRa网络优化方法研究

在农业物联网中，LoRa技术被广泛采用。当网络覆盖范围较大时，标准的LoRaWAN协议会增加扩展因子SF,以增大通信距离。然而，由于SF内(指SF取值相同)和SF间(指SF取值不同)冲突的影响，限制了其节点密度，不适合大范围、高密度的部署场景。为满足大范围和高密度部署要求，结合多跳路由方案和扩频因子分配方法对LoRa网络进行优化，以提高网络的节点容量和可靠性。仿真结果表明，通过使用所提出的方法，在基本节点数量大于10 000个的LoRa网络内，仍可以获得80%以上的数据包交付率PDR,显著提高了大范围、高密度场景下LoRa网络的可靠性。     

基于差异性代码克隆的代码块补全提示方法

在软件编程中,通过代码补全提示可以提高编码效率,但目前缺乏有效的工具和手段从规模差异较大的相似代码中找到合适的候选代码。针对该问题,基于差异性代码克隆技术设计一种新的代码块补全提示方法。通过改进基于滑动窗口和带误匹配索引的匹配算法,寻找与待补全代码块相似的候选代码块,并对其进行特征提取、聚类和相关性排名,以此获得候选代码块的提示顺序。实验结果表明,与HILL等人提出的代码块补全提示方法相比,该方法提示准确率较高,并且适用于更多代码块补全场景。     

Effect of parent selection and sibling rivalry on building-block assembly

Any crossover operator has both beneficial and detrimental effects: it can bring building blocks together or it can tear them apart. In this paper, we provide evidence that the recombination can be biased towards its more beneficial aspects by modifying both the parent selection process and the number of children created by each pair of parents. We exclude both high rank and low rank individuals from being selected as parents. The new idea is that the worst individuals do not have valuable building blocks to contribute, and it is too risky to subject the best individuals to crossover and have their building blocks separated. In a further refinement, we allow the number of children per family to be correlated to the diversity of their parents, and thus increase the pressure of sibling rivalry (competition). These ideas are tested on well-known test functions such as the hierarchical if-and-only-if, royal road, concatenated trap functions and the one dimensional Ising spin model. Four different parent selection schemes are compared and simulations are shown for both two children (fixed) and many children (variable) families. The results indicate that these changes are beneficial for a wide class of problems.     

Scalability problems of simple genetic algorithms

Scalable evolutionary computation has become an intensively studied research topic in recent years. The issue of scalability is predominant in any field of algorithmic design, but it became particularly relevant for the design of competent genetic algorithms once the scalability problems of simple genetic algorithms were understood. Here we present some of the work that has aided in getting a clear insight in the scalability problems of simple genetic algorithms. Particularly, we discuss the important issue of building block mixing. We show how the need for mixing places a boundary in the GA parameter space that, together with the boundary from the schema theorem, delimits the region where the GA converges reliably to the optimum in problems of bounded difficulty. This region shrinks rapidly with increasing problem size unless the building blocks are tightly linked in the problem coding structure. In addition, we look at how straightforward extensions of the simple genetic algorithm-namely elitism, niching, and restricted mating are not significantly improving the scalability problems.     

Moving Block Sequence and Organizational Evolutionary Algorithm for General Floorplanning With Arbitrarily Shaped Rectilinear Blocks

A new nonslicing floorplan representation, the moving block sequence (MBS), is proposed in this paper. Our idea of the MBS originates from the observation that placing blocks on a chip has some similarities to playing the game, Tetris^reg. Since no extra constraints are exerted on solution spaces, the MBS is not only useful for evolutionary algorithms, but also for dealing with rectangular, convex rectilinear, and concave rectilinear blocks, similarly and simultaneously, without partitioning rectilinear blocks into subblocks. This is owed to a special structure designed for recording the information of both convex and concave rectilinear blocks in a uniform form. Theoretical analyses show that the computational cost of transforming an MBS to a floorplan with rectangular blocks, in terms of the number of blocks, is between linear and quadratic. Furthermore, as a follow-up of our previous works, a new organizational evolutionary algorithm (OEA) based on the MBS (MBS-OEA) is proposed. With the intrinsic properties of the MBS in mind, three new evolutionary operators are designed in the MBS-OEA. To test the performance of the MBS-OEA, benchmarks with hard rectangular, soft rectangular, and hard rectilinear blocks are used. The number of blocks in these benchmarks varies from 9 to 300. Also, the MBS-OEA and several well-designed existing algorithms are compared. The results show that the MBS-OEA can find high quality solutions for various problems. Additionally, the MBS-OEA shows a good performance in solving the problems with 300 hard rectangular blocks, 100 soft rectangular blocks, and 100 hybrid blocks, including both soft rectangular and hard rectilinear blocks. This illustrates that the MBS-OEA is not only suitable for solving a wide range of problems, but also competent for solving large-scale problems. Finally, a set of specific experiments is designed to identify the key component that is mainly responsible for the good performance of the MBS-OEA.     

Exploiting the Toeplitz structure in certain queueing problems

Toeplitz matrices model a wide variety of problems in queueing theory, however, the advanced algorithmic tools developed in the field of Toeplitz computations are not generally used in the solution of queueing problems. In this paper we make a survey of some very recent results where concepts as displacement rank, Schur complement, FFT and fast polynomial computations, encountered in Toeplitz matrix analysis, are exploited in order to devise very efficient algorithms for solving a wide class of queueing problems.     

An algorithm to manage variable-length records for highly portable clinical data base systems

An algorithm to archive patient data at free size in disk storage is presented. A record, assumed to be a character string such as an ASCII-coded text, is compressed and divided into fixed-length blocks. One block consists of a data field and a pointer field, and the blocks comprising a record are chained with pointers forwardly. A head pointer of each record is sequentially saved on a separate file. The data compression is performed as follows: if the same character code appears more than twice in succession, we count the number of the repetitions and save it with initial two characters. The algorithms for fetching, re-saving, and purging a record are also presented. These were implemented in FORTRAN77 and tested for performance using a practical patient data file. As the algorithm allows highly flexible record manipulation and can easily be implemented in conventional programming languages, it will make a useful tool for constructing a portable data base management system.     

Dynamic combinatorial optimisation problems: an analysis of the subset sum problem

The field of evolutionary computation has traditionally focused on static optimisation problems. Recently, many new approaches have been proposed that adapt traditional evolutionary algorithms to the dynamic domain to deal with the task of tracking high-quality solutions as the search space changes over time. These novel algorithms are subsequently evaluated on a wide range of different optimisation problems, including well-specified benchmark generators. However, due to a lack of theoretical results, as well as a general lack of references to actual real-world scenarios, it is not entirely clear whether these benchmarks capture any of the characteristics found in NP-hard dynamic optimisation problems. In this paper, we extensively analyse the properties of the NP-hard (dynamic) subset sum problem. In particular, we highlight the correlation between the dynamic parameters of the problem and the resulting movement of the global optimum. It is shown by empirical means that the degree to which the global optimum moves in response to the underlying dynamics is correlated only in specific cases. Furthermore, the role of the representation used to encode the problem, as well as the impact of the formulation of the objective function on the dynamics are also discussed.     

基于Transformer增强架构的中文语法纠错方法

语法纠错任务是自然语言处理领域的一项重要任务,近年来受到了学术界广泛关注。该任务旨在自动识别并纠正文本中所包含的语法、拼写以及语序错误等。该文将语法纠错任务看作是翻译任务,即将带有错误表达的文本翻译成正确的文本,采用基于多头注意力机制的Transformer模型作为纠错模型,并提出了一种动态残差结构,动态结合不同神经模块的输出来增强模型捕获语义信息的能力。受限于目前训练语料不足的情况,该文提出了一种数据增强方法,通过对单语语料的腐化从而生成更多的纠错数据,进一步提高模型的性能。实验结果表明,该文所提出的基于动态残差的模型增强以及腐化语料的数据增强方法对纠错性能有着较大的提升,在NLPCC 2018中文语法纠错共享评测数据上达到了最优性能。     

A block-based evolutionary algorithm for flow-shop scheduling problem

Combinatorial problems like flow shop scheduling, travel salesman problem etc. get complicated and are difficult to solve when the problem size increases. To overcome this problem, we present a block-based evolutionary algorithm (BBEA) which will conduct evolutionary operations on a set of blocks instead of genes. BBEA includes the block mining and block recombination approaches. A block mining algorithm is developed to decompose a chromosome into a set of blocks and rest of genes. The block is with a fixed length and can be treated as a building block in forming a new chromosome later on. To guide the block mining process, a gene linkage probability matrix is defined that shows the linkage strength among genes. Therefore the blocks can be further evolved during the evolutionary processes using this matrix. In the block recombination approach, the blocks along with the rest of genes are recombined to form a new chromosome. This new evolutionary approach of BBEA is tested on a set of discrete problems. Experimental results show that BBEA is very competitive when compared with traditional GA, EA or ACGA and HGIA approaches and it can largely improve the performance of evolutionary algorithm and save a fair amount of computational times simultaneously.     

批次过程控制——回顾与展望

批次过程是一类重要的化工过程.因其本身的灵活性及高效性,被广泛应用于半导体制造、塑料加工、生物制药等领域.针对批次过程控制算法的研究也得到了大批学者的关注.在近三十年中,批次过程控制理论得到了长足的发展.但由于过程本身复杂的动态特性,以及对控制精度要求的提高,现有的理论和方法仍面临着挑战.本文从批次过程的特性出发,分析了算法设计的难点,对几种重要的控制算法进行总结分析,同时讨论了未来可能的发展方向.     

Recent progress in road and lane detection: a survey

The problem of road or lane perception is a crucial enabler for advanced driver assistance systems. As such, it has been an active field of research for the past two decades with considerable progress made in the past few years. The problem was confronted under various scenarios, with different task definitions, leading to usage of diverse sensing modalities and approaches. In this paper we survey the approaches and the algorithmic techniques devised for the various modalities over the last 5 years. We present a generic break down of the problem into its functional building blocks and elaborate the wide range of proposed methods within this scheme. For each functional block, we describe the possible implementations suggested and analyze their underlying assumptions. While impressive advancements were demonstrated at limited scenarios, inspection into the needs of next generation systems reveals significant gaps. We identify these gaps and suggest research directions that may bridge them.