Research on Morlet Wavelet Based Lamb Wave Spatial Sampling Signal Optimization Method

In recent years, Lamb wave and piezoelectric transducers (PZTs) array based wavenumber filtering technique for damage estimation has been gradually studied. Compared with the time domain and frequency domain analysis of the Lamb wave signals, the wavenumber domain analysis is an effective approach to distinguish wave propagating direction and wave modes. However, the spatial resolution sampled by the PZTs is lower than that sampled by scanning laser Doppler vibrometer. As for the diameter of the PZT, it cannot be very small. In this paper, a new Lamb wave spatial sampling signal optimization method based on Morlet wavelet is proposed. Firstly, the frequency band parameter of the Morlet mother wavelet function is calculated by the Lamb wave excitation signal. Then, the sum of squared errors between the Lamb wave spatial sampling signal and the Morlet wavelet function fitting waveform at each scale factor and time factor is calculated. Finally, the scale factor and time factor corresponding to the least sum of squared errors can be judged to be the best match scale factor and time factor respectively, and the Morlet wavelet function fitting waveform in that scale factor and time factor can be seen as the optimized Lamb wave spatial sampling signal. The validation experiment performed on a glass fiber epoxy composite plate shows that the proposed method can improve the spatial resolution and length of the Lamb wave spatial sampling signal, and the sum of squared errors of this method is no more than 0.2.     

Unsupervised spatio-temporal segmentation for extracting moving objects in video sequences

Extraction of moving objects is an important and fundamental research topic for many video ap-plications. This paper addresses an unsupervised spatio-temporal segmentation scheme to extract moving objects from video sequences. The temporal segmentation localizes moving objects by comparing the motion vector of each block in each frame with the corresponding global motion vector estimated by an outlier rejection (OR) based method. Furthermore, the temporal compensation utilizing the temporal coherence of moving objects is considered in the temporal segmentation to solve the temporarily stopping problem. The detected moving regions usually have discontinuous boundaries and some holes. These regions are then compensated in the spatial domain. In the spatial segmentation, the watershed algorithm considering the global information improves the accuracy of segmentation in the spatial domain. The modified mean filter is presented to suppress some minima. By using a fusion module, moving objects are extracted. Experiments on various sequences have successfully demonstrated the validity of the proposed scheme.     

基于加权核密度估计的自适应运动前景检测方法

为解决监控视频背景初始化过程中前景干扰的问题,提出了一种基于加权核密度估计(KDE)的自适应运动前景检测方法.该方法对时间域变化稳定的像素值进行加权,并利用核密度估计构建背景模型,避免了背景初始化过程中前景的干扰.基于该背景模型,提出了一种新的阈值设定策略.该策略根据前景空间分布的连续性自适应获得前景阈值,填充前景中的"孔",并更新阈值.实验结果表明:即使场景中存在运动前景,该方法能够在多种场景下获得90%以上的查准率和查全率,其性能优于传统的背景差法.      

基于差分能量图和CGBP的微表情识别

微表情是一种不能自主控制和伪装的面部表情,其与诚信度的关系密切,具有持续时间短且难以识别的特征.为提高计算机自动识别微表情的准确性,提出一种基于差分能量图和中心化Gabor二值模式（centralized Gabor binary patterns,CGBP）的微表情识别方法.该方法首先利用差分法计算微表情序列的能量得到差分能量图,获得人脸面部肌肉相位的变化;其次将Gabor与中心二值模式CBP相结合,得到CGBP算子对能量图进行微表情的特征提取;最后利用ELM分类器进行微表情分类识别.在CASME微表情库上的实验结果表明,该方法比LBP-TOP、DTSA3、Gabor、VLBP、CBP-TOP算法更能有效地获得微表情序列的时空纹理特征,平均识别率为86.54%.      

城市交通排放高分辨率分析方法研究 ——北京实证

交通污染减排政策呈现多样化、精细化和差别化发展趋势,传统交通排放模型在评价范围、评价尺度上的分辨率存在局限性.本文基于实时监测数据的交通流仿真模型、视频检测方法的车队结构分析和本地化工况的速度排放因子修正关系,提出了城市交通排放高分辨率分析方法,时空分辨率达到小时和路段级别,排放源可区分不同交通方式和本地外地.并以北京市机动车 NOx排放为例,对路网交通排放时空分布特征进行实证分析.北京市高峰时段路网交通排放量占全天排放的31.2%;高速路、快速路排放分别占路网排放总量的37.9%和38.8%;五环~六环间排放量占六环内排放总量的38.32%;货运车辆排放占路网排放比例达到47%.本文提出的城市交通排放高分辨率分析方法对精准定位交通污染治理时间、空间和对象,提升精细化决策水平具有一定意义.     

Research on Multi-Parameter Evaluation of Electric Vehicle Power Battery Consistency Based on Principal Component Analysis

Electric vehicle power battery consistency is the key factor affecting the performance of power batteries. it is not scientific to evaluate the consistency of the battery depending on voltage or capacity. In this paper, multi-parameter evaluation method for battery consistency based on principal component analysis is proposed. Firstly, the characteristic parameters of battery consistency are analyzed, the principal component score can be used as the basis for evaluating the consistency of the battery. Then, the function that multi-parameter evaluation of battery consistency is established. Finally, battery balancing strategy based on fuzzy control is developed. The basic principle of fuzzy control is to fuzzy the input quantity based on expert knowledge, and the fuzzy control quantity is obtained by fuzzy control rule. The results are verified by test.     

考虑客流空间分布的地铁列车节能时刻表优化方法

客流变化引起的列车质量变化是影响地铁列车能耗与节能运行的重要因素之一. 本文考虑地铁线路客流空间分布差异，研究耗散型再生制动能利用方式下的列车节能时刻表优化方法. 结合各区间载荷差异和列车运动方程，建立以净能耗最小为目标的时刻表优化模型，通过适度优化计划停站时间、区间运行时间和折返时间协同多列车牵引、巡航、惰行和制动过程的时空分布，设计二分法和粒子群算法对模型求解. 以北京地铁某线路进行实例研究，结果表明，优化模型能有效协同多列车的节能运行，考虑客流空间分布差异比假定列车载荷为常数能进一步提升节能效果.     

基于时空行为规律挖掘的公交乘客分类方法

应用智能公交系统(APTS)提取个体乘客出行信息，构造了公交出行链，研究了基于时空行为规律挖掘(STBRM)的公交乘客分类方法；应用时间序列表征乘客出行时间特征，利用互相关距离(CCD)算法计算了个体乘客出行时间规律；应用带噪声基于密度的空间聚类(DBSCAN)算法，挖掘了个体乘客的出行空间规律；依据出行强度和出行时空规律，将乘客划分为极少出行、时间规律、空间规律、时空规律和不规律等5个群体；以出行天数、类似上车时间数量和类似上车站点数量为聚类指标，应用K-Means++算法将乘客划分为高规律、中规律和低规律3类，比较了本文提出的STBRM方法和K-Means++聚类方法的分类结果，揭示了2种方法分类结果之间的关系。研究结果表明: 当时段划分长度取5 min，时间规律性判断阈值取3.0时，利用CCD算法识别时间模式规律乘客的效果最佳，与常用的DBSCAN算法相比，识别率提升了14.64%；增加时间窗长度能够提高时间、空间模式规律判定结果的稳定性；时间窗长度达到3周后，空间模式规律的乘客比例下降趋缓，达到6周后趋于稳定；时间窗长度达到2周后，时间模式规律的乘客比例增长趋缓，达到4周后趋于稳定；时间规律、空间规律和时空规律等3类乘客数量仅占总乘客数量的30.4%，但其出行量占到了总出行量的84.7%，公交依赖度很高，应作为公交机构重点保障的对象；本文提出的STBRM方法与K-Means++聚类方法的分类结果具有较强的关联性，规律性极高或极低的群体高度重合。      

New Method of Reconstruction from Nonparallel Stereo and Application to Surgical Navigator

A new method to reconstruct 3D scene points from nonparallel stereo is proposed. From a pair of conjugate images in an arbitrarily configured stereo system that has been calibrated, coordinates of 3D scene points can be computed directly using the method, bypassing the process of rectifying images or iterative solution involved in existing methods. Experiment results from both simulated data and real images validate the method. Practical application to surgical navigator shows that the method has advantages to improve efficiency and accuracy of 3D reconstruction from nonparallel stereo system in comparison with the conventional method that employs algorithm for standard parallel axes stereo geometry.     

移动荷载与土体中孔洞相互作用的2.5D边界元法分析

为了研究具有复杂表面的饱和土体与移动荷载相互作用机理,根据饱和土Biot理论,采用Fourier变换和势函数分解法,推导了饱和土体频域-波数内的Green函数;建立了移动荷载作用下,饱和土体中孔洞动力响应的频域-波数域内边界元法方程(2.5D boundary element methods);利用快速Fourier逆变换法,得到了时间-空间域内饱和土体的动力响应.研究结果表明:在频域内,利用移动荷载方向的一致性建立的频域-波数域内边界积分方程,可将3D空间问题转化为频域-波数域的2D平面问题, 3D边界元简化为2D边界元,使得计算面转换为计算线,减小了计算规模.      

A Freeway Exit Volume Prediction Method Based on Traffic Origin-Destination Stability Pattern

The traffic origin-destination (OD) data can be directly obtained from the freeway toll data. This information always reflects the vehicle status in temporal and spatial. And the OD information can also be used for making reasonable and effective predictions on the future trends of the freeway entrance volume. In this paper, an algorithm is proposed for predicting the freeway entrance volume based on the analysis of stability pattern of freeway traffic OD data. The contributions mainly include: (1) An information entropy-based method is presented for analyzing stability pattern of freeway traffic OD data; (2) Based on the stability pattern, a new prediction algorithm is applied on massive freeway entrance data. (3) The experimental results show that the proposed algorithm is reasonable and feasible, the accuracy is improved as well.     

一种基手双目线结构光视觉系统的简单标定方法

提出了一种基于双目线结构光主动三角法的视觉系统的简单标定方法．该视觉系统采用线结构光原理,使用双目摄像机避免了常用的单摄像机系统中容易出现的遮挡等问题．该系统的标定综合了线结构光传感器模型以及双目立体视觉模型,推导了各个标定坐标系之间的联系,依据蔡氏两步法,使用方格靶标和最小二乘法求得各种标定参数．实验结果表明,该标定方法操作简单、实用可行．     

群决策中基于SVDD的专家权重的评价方法

针对群决策中专家权重的评价问题,提出一种基于支持向量域描述(SVDD)的确定方法.利用生成树的方法把判断矩阵进行一致性剖分,利用支持向量域描述的方法排除干扰信息,找出群体公共信息,并引入群体相容性、群体信息贡献率等概念及其判断方法;给出不同专家的群体信息贡献率的计算方法,确定出每位专家的评价权重,并通过示例验证了该方法的可行性和有效性.该方法能消除片面的影响,抓住关键信息,对专家的权重能比较客观地评价.     

����OD�ȶ�ģʽ�ĸ��ٹ�·��������Ԥ�ⷽ���о�

从高速公路车辆出入口数据中可以直接提取路网中车辆的OD信息,这些信息在一定程度上反映出车辆时空上的运行状态,更可以通过这类数据对高速公路出口流量的趋势做出合理有效地预测.本研究从真实数据出发,对交通路网中车辆OD的稳定性进行了分析,并以此为指导,将所得到的领域知识加入到常规的预测方法中,提出了一套基于OD稳定模式的高速公路出口流量预测方法.主要贡献包括：①提出了信息熵的概念,以刻画高速公路车辆OD的规律性;②提出了基于OD的稳定模式的出口流量预测方法;③在大量真实的数据上进行了实验分析.     

基于调制白噪声与查表法的非平稳路面不平度建模方法

针对实测道路功率谱密度频率结构不同与汽车行驶速度变化引起的非平稳问题, 分析了现有路面不平度模型的准确性, 提出了一种基于调制白噪声与查表法的非平稳路面不平度建模方法; 结合白噪声生成方法和离散功率谱密度的计算公式, 推导出了能够确保路面不平度计算功率谱与设计功率谱完全吻合的Band-Limited White Noise模块参数设定方法; 通过不同设定值组合的白噪声功率谱密度与路面不平度功率谱密度计算结果对比, 验证了所提出的Band-Limited White Noise模块参数设定值的准确性; 为解决频率指数不等于2的空间域内非平稳路面不平度建模问题, 提出了由1个基准项和若干个精度校正项组成, 且调制传递函数模的平方逼近具有任意频率指数的道路功率谱表达式; 为获取考虑车速变化影响的频率时变的时域内非平稳路面不平度输入模型, 基于空间路面不平度固定不变的原理, 提出了利用查表法, 获得变车速工况下时域内非平稳路面不平度; 通过空间域内非平稳路面不平度建模及考虑车速变化的非平稳路面不平度建模应用, 验证了所提建模方法的优越性。研究结果表明: 使用基于调制白噪声的非平稳路面不平度建模方法构建的空间域路面不平度功率谱曲线, 与应用实例中的设计目标曲线吻合良好, 且结合查表法可以有效地解决时变车速带来的时域内非平稳路面不平度建模问题。      

自适应铁路场景前景目标检测

现代铁路系统中，智能视频分析技术已被广泛应用于异物入侵监测，前景目标检测是入侵判断的必要过程. 背景差分常用于检测前景目标，但铁路场景复杂，存在动态变化的背景区域和未知类型的目标，现有基于阈值分割或深度学习的背景差分算法都不能满足需求，故提出一种基于阈值自适应调节的前景目标检测算法. 利用像素值在时间上的动态信息，分割结果的反馈信息和由超像素提供的空间信息确定阈值调节因子，动态调节阈值以适应环境变化；提出一种灵活可靠的背景模型初始化方法，消除鬼影问题，实现一帧到多帧初始化的灵活切换. 实验结果表明，所提算法在铁路场景上取得了较好的准确率和误分类率，且平衡了精度和速度.     

基于二维浅水方程的直线段沥青路面径流特性

基于二维浅水方程的水动力学方法建立了直线段沥青路面径流的数值模型, 根据实际降雨条件下沥青路面径流变化过程的监测结果验证了模型参数, 研究了路面宽度、组合坡度等几何参数与路侧排水方式对路面径流时空分布特性的影响。研究结果表明: 设计降雨条件下, 路面径流在空间分布上呈较强的二维特性, 沥青路面径流深度变化依次经历增加、稳态径流与退水3个过程; 漫排水条件下, 路面宽度分别为11、15、20、25、30 m时, 路面径流最大深度分别为11.87、14.39、17.08、19.69、21.98 mm, 退水时间分别为1.4、1.4、2.4、2.9、3.4 min; 路面径流深度增幅随路面宽度的增加而降低, 退水时间随路面宽度的增加而增加; 相比于行车道, 硬路肩路面径流的退水时间延长约20%;较大的坡度组合(横坡为3%, 纵坡为2%) 有利于排水; 当采用集中排水时, 路缘石的阻拦使路侧产生壅水, 壅水区宽度为6~8 m, 壅水区范围占路面宽度的比例随路面宽度的增加而逐渐缩小, 非壅水区内的路面径流深度变化与漫排水条件下基本相同; 为保证行车安全, 可通过改变路面坡度来减少路面径流的汇流时间; 路缘石对路面径流的阻拦效应明显, 在排水设计中应合理设置路缘石高度与开口间隔, 避免行车道出现壅水现象。      

基于深度学习的网约车供需缺口短时预测研究

城市不同区域网约车供需缺口预测可为车辆调度策略提供支持,从而提高车辆运行效率和乘客服务水平.为实现网约车供需缺口短时预测,提出一种基于时空数据挖掘的深度学习预测模型(Spatio-Temporal Deep Learning Model, S-TDL).该模型由时空变量模型、空间属性变量模型和环境变量模型 3个子模型融合而成,可捕捉时空关联性、区域差异性和环境变化对供需缺口的影响.同时,提出特征聚类-最大信息系数两阶段特征选择方法,筛选与供需缺口相关性强的特征变量,提高训练效率,减少过拟合.滴滴出行实例分析证明,特征选择后的 STDL模型预测精度显著优于BP神经网络、长短期记忆网络和卷积神经网络.     

Recognition Algorithm and Risk Assessment of Airport Hotspots

The hotspot recognition algorithm is proposed based on a potential collision in order to study the aircraft taxi conflicts in large airports. The spatial and temporal distribution characteristics of hotspots are analyzed based on the risk assessment model of hotspot constructed in this paper. Firstly, approaches for monitoring of the aerodrome movement were compared. The hotspot recognition algorithm taken into account of whether aircrafts' taxi track has spatial and temporal overlap based on the aerodrome surveillance radar(ASR) data was presented, by identifying the hotspots through analyzing whether the aircrafts' time of entering and exiting the same taxiway is overlap or not, and the heading difference and distance of the two aircrafts satisfy the specified threshold constraint condition. Then, the ASR data were divided into several parts, and then airport hotspots were recognized and the spatial and temporal distribution characteristics were analyzed. The risk assessment model of airport safety hotspots was constructed which is taken into account of the conflict probability and its severity consequence. Finally, based on the risk grade assessment criteria and hotspots' risk value, the risk grade ranking of hotspot in one airport of China was evaluated and designated. According to the result, the spatial and temporal distribution characteristics of airport hotspots were varied with the variation of airport traffic flow and operational mode of runway, which shows that the hotspots have the characteristics of dynamic periodicity and diurnal variation. And the risk assessment results were consistent with experts' opinions and actual operation condition, which verified the rationality of the hotspot recognition algorithm, risk assessment model as well as the risk grade ranking criteria.