A global energy optimization framework for 2.1D sketch extraction from monocular images

The 2.1D sketch is a layered image representation, which assigns a partial depth ordering of over-segmented regions in a monocular image. This paper presents a global optimization framework for inferring the 2.1D sketch from a monocular image. Our method only uses over-segmented image regions (i.e., superpixels) as input, without any information of objects in the image, since (1) segmenting objects in images is a difficult problem on its own and (2) the objective of our proposed method is to be generic as an initial module useful for downstream high-level vision tasks. This paper formulates the inference of the 2.1D sketch using a global energy optimization framework. The proposed energy function consists of two components: (1) one is defined based on the local partial ordering relations (i.e., figure-ground) between two adjacent over-segmented regions, which captures the marginal information of the global partial depth ordering and (2) the other is defined based on the same depth layer relations among all the over-segmented regions, which groups regions of the same object to account for the over-segmentation issues. A hybrid evolution algorithm is utilized to minimize the global energy function efficiently. In experiments, we evaluated our method on a test data set containing 100 diverse real images from Berkeley segmentation data set (BSDS500) with the annotated ground truth. Experimental results show that our method can infer the 2.1D sketch with high accuracy.     

基于Flash 3D的在线虚拟漫游系统设计与实现

对目前虚拟漫游系统的瓶颈进行分析,提出并实现了一个基于Flash 3D的、优化了的在线虚拟旅游系统。系统对三维场景文件包括模型文件、材质文件及其构建步骤进行优化,并在实现了对多种媒体元素支持的基础上,设计了独创的热点系统,实现对三维场景按需进行加载和展示,以及游客在场景中的互动漫游。系统测试结果表明,该设计方案优化效果显著,系统性能得到了极大的提升。     

一种融合多帧ICP和图优化的算法研究

当前基于迭代最近点拼接的同时定位与建图算法,存在误差积累、无法满足大范围定位精度的缺陷。为此,提出一种融合多帧迭代最近点和图优化的算法。在时域上处理点云拼接问题,将单帧迭代最近点算法推广到多帧进行最近点迭代,提取同一地点在不同时刻的数据特征,形成多个封闭循环,再运用基于最小二乘的图优化方法对点云拼接后的全网数据进行全局优化,消除累计误差,提升整体的定位精度。采用鲁巷和密歇根的数据进行测试,结果表明,该方法在一定程度上减少了匹配误差,平均误差为1.0m,最小误差为0.2m,可以满足大范围同步定位与建图的精度需求。     

基于QPSO 算法移动机器人轨迹规划与实验

针对移动机器人路径规划问题,提出一种基于QPSO算法的路径规划方法,并用概率论的方法分析了移动机器人路径规划的收敛性,阐明了该方法随均匀分布和正态分布的参数关系和收敛区间;然后根据移动机器人的运动特征提出一种改进的轨迹规划方法。移动机器人平台的实验结果表明了该方法在移动机器人路径规划中的有效性和可行性。     

基于梯度投影法与随机优化算法的约束优化方法

针对带有线性等式和不等式约束的无确定函数形式的约束优化问题,提出一种利用梯度投影法与遗传算法、同时扰动随机逼近等随机算法相结合的优化方法。该方法利用遗传算法进行全局搜索,利用同时扰动随机逼近算法进行局部搜索,算法在每次进化时根据线性约束计算父个体处的梯度投影方向,以产生新个体,从而能够严格保证新个体满足全部约束条件。将上述约束优化算法应用于典型约束优化问题,其仿真结果表明了所提出算法的可行性和收敛性。     

基于A-VFSA粒子发生器的PSO算法的研究

对超快速模拟退火算法进行改进(A-VFSA),并以此为粒子发生器,提出了PSO-PG算法.该算法通过粒子发生器对搜索域内粒子进行改进,从而生成一个精英粒子池;并根据随机策略从粒子池中选择粒子,作为初始种群,采用PSO算法进行优化,得到全局最优解.相比于标准PSO算法和LDW算法,PSO-PG算法拥有更好的稳定性和优化精度,能够更加快速地收敛到全局最优解,在一定范围内几乎不依赖于初始参数的选择.     

多周期多原料的供应链网络成本优化及其混合遗传算法

建立并求解一个基于成本最小的供应链网络模型.与以往研究不同,在该模型中生产一种产品需要至少两种原料,每种原料都可以由备选供应商提供.根据模型的特点,用0、1代表对原材料供应商、工厂和分销中心的选择情况,以MATLAB 7.6为平台,运用Sheffield大学的遗传算法工具箱,将遗传算法与线性规划算法相结合,实现了模型的求解.算例结果表明,给出的染色体编码方案正确,混合遗传算法有效,能解决多周期、多原料的供应链网络成本优化问题.还探讨了需求和距离变化,以及需求随机时对最优成本和最优个体的影响.研究表明,需求变化的影响大于距离变化的影响,需求随机对最优成本和最优个体的影响不大.     

文本分类中训练集相关数量指标的影响研究

针对训练集对分类性能的影响,从训练集的文本数、类别数以及特征项数这三项数量指标出发进行研究。使用多因素方差分析方法及多种语料库定量探讨该三项数量指标对分类性能的影响规律。结果发现特征项数对分类性能的影响在不同的文本数和类别数时是不同的,分类性能受训练集的这三项指标的交互影响,通过对训练集的这三项指标进行优化,提出了从分类算法、特征项选择法以外提高分类性能的途径。在真实数据上的实验结果表明,该方法可有效提高分类性能。     

An analysis,design and tuning of Cascade Control Systems in the presence of constraints in actuator and process outputs

In the presence of rate constraints in actuator, design of cascade controller based on the primary controller conditional integration can result into closed-loop system oscillations. Stability analysis, performed by the describing function technique and confirmed by simulation, demonstrates that the solution based on the Anti-Reset Windup Cascade Control System (ARW CCS) structure is successful. Design and tuning of the ARW CCS secondary controller is a standard ARW single-input single-output problem. In the present paper tuning is proposed for the ARW CCS primary controller. For the serial process modeling simple rules are derived and confirmed by experimental results, obtained on a drum type boiler of a 210 MWe lignite coal fired unit. General design of the ARW CCS is based on the parallel process modeling and optimization of the primary controller. Optimization is performed in the frequency domain, under constraints on the maximum sensitivity, multiplicative uncertainty bound and sensitivity to measurement noise. Simulation and experimental results on a laboratory thermal plant demonstrate effectiveness of the proposed optimization.     

A bi‐objective approach to reschedule new jobs in a one machine model

We consider a one machine scheduling model, minimizing a classical objective function—either the total completion time or the maximum tardiness—and with two sets of jobs: one with initial jobs already scheduled and one with new jobs that must be inserted in the schedule. As such rescheduling can create a modification of the schedule of the initial jobs, a disruption objective is considered in addition to the original objective. This additional objective can be formulated in four different ways. Such model has been introduced by Hall and Potts, minimizing either a linear aggregation of the two objectives or the initial objective under a constraint giving an upper limit of the disruption objective. In this paper, the aim is to obtain the set of efficient schedules in regard to the two objectives. Algorithms are provided for the eight possible bi‐objective problems and illustrated by some didactic examples.