Two-Staged Method for Ice Channel Identification Based on Image Segmentation and Corner Point Regression

Identification of the ice channel is the basic technology for developing intelligent ships in ice-covered waters,which is important to ensure the safety and economy of navigation.In the Arctic,merchant ships with low ice class often navigate in channels opened up by icebreakers.Navigation in the ice channel often depends on good maneuverability skills and abundant experience from the captain to a large extent.The ship may get stuck if steered into ice fields off the channel.Under this circumstance,it is very important to study how to identify the boundary lines of ice channels with a reliable method.In this paper,a two-staged ice channel identification method is developed based on image segmentation and corner point regression.The first stage employs the image segmentation method to extract channel regions.In the second stage,an intelligent corner regression network is proposed to extract the channel boundary lines from the channel region.A non-intelligent angle-based filtering and clustering method is proposed and compared with corner point regression network.The training and evaluation of the segmentation method and corner regression network are carried out on the synthetic and real ice channel dataset.The evaluation results show that the accuracy of the method using the corner point regression network in the second stage is achieved as high as 73.33%on the synthetic ice channel dataset and 70.66%on the real ice channel dataset,and the processing speed can reach up to 14.58frames per second.

Fine-grained grid computing model for Wi-Fi indoor localization in complex environments

The fingerprinting-based approach using the wireless local area network(WLAN)is widely used for indoor localization.However,the construction of the fingerprint database is quite time-consuming.Especially when the position of the access point(AP)or wall changes,updating the fingerprint database in real-time is difficult.An appropriate indoor localization approach,which has a low implementation cost,excellent real-time performance,and high localization accuracy and fully considers complex indoor environment factors,is preferred in location-based services(LBSs)applications.In this paper,we proposed a fine-grained grid computing(FGGC)model to achieve decimeter-level localization accuracy.Reference points(RPs)are generated in the grid by the FGGC model.Then,the received signal strength(RSS)values at each RP are calculated with the attenuation factors,such as the frequency band,three-dimensional propagation distance,and walls in complex environments.As a result,the fingerprint database can be established automatically without manual measurement,and the efficiency and cost that the FGGC model takes for the fingerprint database are superior to previous methods.The proposed indoor localization approach,which estimates the position step by step from the approximate grid location to the fine-grained location,can achieve higher real-time performance and localization accuracy simultaneously.The mean error of the proposed model is 0.36 m,far lower than that of previous approaches.Thus,the proposed model is feasible to improve the efficiency and accuracy of Wi-Fi indoor localization.It also shows high-accuracy performance with a fast running speed even under a large-size grid.The results indicate that the proposed method can also be suitable for precise marketing,indoor navigation,and emergency rescue.

Combination Operation for Neutral Point of 35 kV Power Grids in Mountainous Area

Based on a lot of measurement and analysis,this paper find that the unbalanced nature of 35 kV grids due to unsymmetrical capacitance values gives difficulty in neutral point operation in mountainous area.Improving compensation of Petersen coil can rectify imbalance of voltage but bring up new problem that Petersen coil can not extinguish grounding arc effectively in fault.To put down contradiction mentioned above,this paper proposes a combination operation for neutral point of 35 kV grids as neutral point insulated in routine operation and grounding through Petersen coil in fault,then EMTP simulation is carried out.Simulation indicates that new neutral operation method can improve lightning withstand level and decrease trip-out rate of grids.

PLC-Oriented Access Point Location Planning Algorithm in Smart-Grid Communication Networks

In this study,we investigate the optimal location of access points(APs) to connect end nodes with a service provider through power-line communication in smartgrid communication networks.APs are the gateways of power-distribution communication networks,connecting users to control centers.Hence,they are vital for the reliable,safe,and economical operation of a power system.This paper proposes a planning method for AP allocation that takes into consideration economics,reliability,network delay,and(n-1) resilience.First,an optimization model for the AP location is established,which minimizes the cost of installing APs,while satisfying the reliability,network delay,and(n-1) resilience constraints.Then,an improved genetic algorithm is proposed to solve the optimization problem.The simulation results indicate that the proposed planning method can deal with diverse network conditions satisfactorily.Furthermore,it can be applied effectively with high flexibility and scalability.

On REE and EER Methods for Mining Corner Points on the Images

This paper reviews, implements and compares two corner detection algorithms for mining corner points on the generic shapes. These corner detectors detect corners by using combination of one rectangle (R) and two ellipses (EE). These algorithms have been used with different combinations: REE and EER together with different parameter settings in their descriptions. REE and EER combinations slide along the boundary of the shape and record number of boundary points in each rectangle and ellipses. REE and EER setup represent both local and global views of the image outlines and present natural corner detection methodologies to detect and mine all true corners accurately. A comparative study demonstrates the superiority of the REE and EER over some of the existing algorithms.

THE DERIVATION OF EXACT STATIC CONDITIONS AT THE CORNER POINTS FOR THE BENDING OF THICK RECTANGULAR PLATES

In this paper, exact static conditions at the corner points for the bending of thickrectangular ptates are strictly. derived from the theorem of minimum potentialenerg[1].

Mining Corner Points on the Generic Shapes

This paper designs and implements a corner detection algorithm for mining corner points on the generic shapes. The proposed corner detector detects corners by using combination of one rectangle and two ellipses (REE) with different parameter settings in their descriptions. REE combination slides along the boundary of the shape and records number of boundary points in each rectangle and ellipses. REE setup represents both local and global views of the image outline. The proposed technique presents a natural corners detection methodology to detect all true corners accurately. This technique is consistent with human vision system.

VIBRATIONS OF RECTANGULAR PLATES SUPPORTED AT CORNER POINTS

This paper discusses by energy theorem the methodof approximate computation for the lowest eigenfrequencies of rechmguhir plates,on which there are symmetrical concentrated masses,supported at corner points,In the case of seseral concentrated masses,by using the prineiple of superposition we mayfiml the reduneed coefficients of masses comveniently.llence we can louain the lowest eigenfrequencies of thin plates.In the paper a good mamy mmerical caleuhting eximples are inustrated

融合角点检测的点云配准研究

为提高点云配准效率,提出一种基于点云曲率信息的Harris角点提取算法,通过引入曲率约束,对传统Harris角点提取做出改进;计算每个点的曲率值并与整幅点云的平均曲率作比较,保留曲率值大于平均曲率的点云,实现对配准价值较低的点的剔除,为后续点云配准减少计算量;构建所提取角点之间的特征匹配,结合采样一致性(sample consensus initial alignment, SAC-IA)粗配准和正态分布变换(normal distributions transform, NDT)精配准,寻找最优的变换矩阵,实现源点云和目标点云的重合。通过比较不同算法在公开数据集上进行实验验证,结果表明该方法对复杂点云、多拐点点云都有较好的表现,对初始位置相差较大的点云也有良好的适用性。

非结构化网格下海洋流场的特征提取与种子点选取算法

对于非结构化网格流场数据,现有研究存在不能定位流场中临界点的具体位置、无法对临界点精细分类以及缺乏适用于非结构化网格的种子点选取算法等问题。针对上述问题,基于非结构化网格流场数据,分别在临界点提取和种子点选取方法上进行改进创新。提出非结构化网格中临界点的定位及分类方法,通过庞加莱指数法判断存在临界点的三角网格。构造质心迭代法定位临界点在网格中的准确位置,并设计三角网格内雅克比矩阵的构造方法,将临界点精细分类。基于非结构化网格提出基于最大得分和网格密度的种子点选取算法,先比较相邻格点的标量值大小来计算每个格点的得分,形成“最大得分”标量场,再按照网格密度动态设置阈值,将得分大于阈值的格点选为种子点,接着以种子点为起始点生成流线,生成的流线可以表达出流场的关键特征与全局信息。基于多个海域流场数据的实验结果表明,临界点分类的准确率可达99%以上,证明了临界点提取算法的准确性以及种子点选取算法对提升流场可视化效果的有效性。

三层正放四角锥预应力网架稳定性分析

网架以其优良的工程性能成为大跨度结构的首选结构类型之一。为改善传统网架结构在大跨度条件下变形大、用钢量大的缺陷,文中通过对网架施加预应力的方式,达到增强结构整体刚度和承载能力的目的。预应力的施加必然会对网架的稳定性能造成影响,其中支承形式对网架结构的稳定性影响较大。并通过对网架施加不同预应力后进行稳定性分析,得到不同预应力布索方式对网架稳定性影响,为网架的实际工程建设提供参考。

融合栅格与表面特征编码的轻量级点云分类网络

点云携带着丰富的三维特征,其分类问题一直是深度学习领域的热点。现有点云分类网络的精度已经较为理想,但参数量与计算量过于庞大,不利于实际场景下的部署。针对该问题,提出一种轻量级点云分类网络GridPoint。首先设计点云栅格化模块Gridding,根据点的坐标位置进行栅格区域划分;然后扩展坐标的高次项函数,对原始点云进行表面特征编码,增强对轮廓特征的表达;最后使用两次全局池化完成局部特征的提取与全局特征的聚合。使用经典点云数据集ModelNet40、ShapeNetCore与真实数据集ScanObjectNN进行分类与消融实验。实验结果表明,GridPoint的分类精度接近PointNet++等主流网络,差距在0.3%~2.3%之间;网络参数量与计算量分别为0.11 M与0.05 G,相较主流网络分别减少了81.7%和88.9%以上,在轻量化方面优势显著,具有良好的实用价值。

计及光伏波动性的主动配电网双层有功无功协调优化

针对含有高比例分布式电源的配电网在源荷不确定性和协调多种无功补偿设备等方面存在的问题,提出一种主动配电网双层有功无功协调优化策略。该策略首先针对离散型出力的有载调压变压器和电容器组,以网损和电压偏差为目标,建立上层优化模型;其次,下层优化模型通过储能出力平抑光伏的有功波动,同时通过调整静止无功补偿装置、光伏逆变器弥补光伏并网点的无功缺额,以配电网的有功波动和光伏并网点的电压偏差为目标,并采用改进的多目标粒子群优化算法求解该模型;最后,通过IEEE 33节点算例验证所提策略的有效性。该优化策略不仅提高了有功与无功的协调性,且实时性强,不依赖精准的设备控制,使配电网的经济性和可靠性得到提升。

一种基于激光点云数据的微距栅格体积算法

针对规则格网算法难以满足激光点云模型高精度体积计算的问题,提出了一种基于激光雷达点云数据的微距栅格体积算法。该方法首先运用葛立恒凸包算法提取凸包点集,然后运用微距格网划分、高程插值和网格体积累加的方法计算体积。与规则格网法不同,这种算法充分利用激光雷达数据高密度点云特征,采用格网微分和增大插值半径的方法改善模型表面的连续性,进而提高计算精度。实验结果表明,微距栅格体积算法具有较好的时间复杂度和较高的计算精度,适宜于激光点云模型高精度体积计算。

基于鱼眼相机混合模型的棋盘格角点检测方法

针对搭载自动泊车功能车辆的鱼眼相机标定耗时较长、鲁棒性要求较高的问题,提出一种基于鱼眼相机混合模型的棋盘格内角点检测改进方法。在参考和比较原有算法的基础上,提出的混合畸变模型能够快速确定不同区域内角点的两个初始方向,从而快速提高粗定位的速度。同时针对生产线标定的特殊性,提出了双标准离率差阈值作为对初始棋盘格是否可用的判断标准。实验结果表明该算法在自动泊车鱼眼相机生产过程中既能够满足标定的精度要求,同时也大幅度提高了检测速度,对自动泊车的相机标定具有一定的实用价值。

基于Grid-GSA算法的植保无人机路径规划方法

为了提高植保无人机的作业效率,研究了一种路径规划方法。运用栅格法构建环境模型,根据实际的作业区域规模、形状等环境信息和无人机航向,为相应栅格赋予概率,无人机优先选择概率高的栅格行进。基于上述机制实现了在形状不规则的作业区域内进行往复回转式全覆盖路径规划;以每次植保作业距离为变量,根据仿真算法得出返航点数量与位置来确定寻优模型中的变量维数范围,以往返飞行、电池更换与药剂装填等非植保作业耗费时间最短为目标函数,通过采用引力搜索算法,实现对返航点数量与位置的寻优;为无人机设置必要的路径纠偏与光顺机制,使无人机能够按既定路线与速度飞行。对提出的路径规划方法进行了实例检验,结果显示,相比于简单规划与未规划的情况,运用Grid-GSA规划方法得出的结果中往返飞行距离总和分别减少了14%与68%,非植保作业时间分别减少了21%与36%,其它各项指标也均有不同程度的提高。在验证测试试验中,实际的往返距离总和减少了322 m,实际路径与规划路径存在较小偏差。验证了路径规划方法具有合理性、可行性以及一定的实用性。

一种顾及时空变化的BDS PPP多路径误差建模方法

根据BDS GEO、IGSO、MEO三种星座卫星轨道运行特性,使用多路径误差半天球格网点模型(multi-point hemispherical grid model,MHGM)和恒星日滤波(sidereal filtering,SF)方法建立BDS混合星座的多路径误差改正模型。使用该模型改正多路径误差后,BDS精密单点定位(precise point positioning,PPP)的载波相位验后残差显著减小,E、N、U方向定位精度分别提升41%、37%、38%,收敛速度整体提升31%。

自由节点差分法寻点策略研究及验证

混合网格具有良好的贴体性,适用于Navier-Stokes方程的计算,但在网格交界面处需要通过插值进行流场信息交换,而这一过程会引入误差。陈洁等提出了一种适用于无序网格点的自由节点差分计算方法,该方法可对重叠网格交界面流场进行差分计算,无须采用插值方法进行流场信息传递,解决了插值方法引入误差的问题。但该自由节点差分法的计算模板需要从中心点周围的点云中选择网格点构成,不同选点策略对计算结果的影响不同。针对自由节点差分法的选点需求,本文综合考虑角度、正交性、距离等因素,提出了几种不同选点策略,并利用数值实验进行了验证。结果表明,本文提出的选点策略符合自由节点差分法的构造思想,均能获得稳定的收敛解,其中模板点与中心点距离对计算精度影响最大,是选点策略的首要影响因素。

基于网格的多点在多边形内高效判定方法

提出一种基于网格的点(多点)在多边形内高效判定方法。预处理阶段使用数值微分法(DDA)结合边界代数法,快速识别边界网格并分割出多边形边片段,同时标记网格左下角点位置属性;当待判定点位于非边界网格内时,根据预处理结果直接判定;位于边界网格时,对边界网格分块后再判定。实验结果表明,该预处理方法高效快速且边界网格分块法有效缩减了判定时间。此方法适用于凹凸多边形、自相交多边形以及环状多边形,相较现有算法优势明显,100万个点在28012条边的多边形中判定用时约0.04 s。

地面激光雷达扫描参数与点云简化对林木参数提取的影响

【目的】林木参数是森林蓄积量、森林生物量估算的基础指标,传统的人工调查方式费时费力,已难以适应新形势下数字化森林资源监测技术的要求。地面激光雷达扫描技术能够获取小尺度高分辨率的林分内部结构信息,为林分环境条件下林木胸径、树高提取提供一种新的思路。【方法】以芦头实验林场杉木林样地为研究对象,针对FARO Focus 3D X330三维激光扫描仪设计了7种不同的扫描组合方式对样地进行扫描,提出象限角点云简化思路进行参数提取和精度评价,探究不同扫描组合方式对林木胸径、树高参数提取精度与效率的影响。【结果】1)当扫描分辨率为1/2、质量为4X时,胸径参数提取精度最高;当扫描分辨率为1/4、质量为4X时,树高参数提取精度最高。2)在林木参数提取结果没有显著性差异的前提下,扫描分辨率为1/4、质量为4X的扫描参数工作效率最高。3)选取同时兼顾精度和效率的1/4扫描分辨率、质量4X的扫描结果,进行象限角点云简化,简化的点云能够准确地提取出林木胸径参数。【结论】研究结果对于具有相同或相似地理条件和树种的林地选择扫描参数和点云简化方式具有重要参考价值,可以提高内业工作效率,同时也为地面激光雷达野外样地调查提供方法和技术参考。