激光雷达和相机的决策级融合目标检测方法

【目的】激光雷达与相机这两类传感器检测数据格式不统一、分辨率不同,且数据级和特征级的融合计算复杂度高,故提出一种决策级的目标融合检测方法。【方法】对激光雷达与相机的安装位置进行联合标定,实现这两类传感器检测结果的坐标系转换;利用匈牙利算法将激光雷达点云检测目标框和相机图像检测目标框进行匹配,设定目标框重合面积阈值,检测获得目标物的位置、类型等。【结果】实车测试结果表明,根据检测目标检测框长宽比选取不同交并比阈值的方法使得车辆和行人的目标识别准确率分别提升了3.3%和5.3%。利用公开数据集KITTI对所提融合方法进行验证,结果表明,在3种不同难度等级场景下,所提融合方法的检测精度分别达到了75.42%、69.71%、63.71%,与现有常用的融合方法相比,检测精度均有所提升。【结论】这两类传感器的检测目标框重合面积阈值对决策级融合检测结果影响较大,根据检测目标检测框长宽比选取不同阈值可有效提升车辆和行人的目标识别准确率。决策级融合方法能准确匹配雷达和相机的检测目标,有效提升目标检测精度。

协同上下行链路的频谱资源分配方案

传统蜂窝网络采用固定的频谱资源分配方式,导致频谱利用率低且无法满足用户高速和高质量通信需求。针对这种局限性提出了一种将分布式天线系统(Distributed Antenna System,DAS)与终端直通(Device to Device,D2D)通信技术相结合的策略。首先,该方案提出协同上下行链路(Uplink and Downlink Collaboration,UADC)的资源分配算法,并构建了以最大化系统总效率为目标的非线性规划问题。然后,将该问题分为最佳功率选择和信道分配2个子问题,同时利用匈牙利算法为D2D对选择最佳信道。最后,在MATLAB仿真平台进行实验。仿真结果表明,相比于仅复用上行链路的和仅使用DAS系统的方法,所提方案的频谱效率更高。

利用改进匈牙利算法求解旅行商问题

针对传统的匈牙利算法在求解旅行商问题(travelling salesman problem,TSP)时会导致多回路闭合的问题,提出了破环机制,设计了破环匈牙利算法。通过采用分配问题的描述方法对旅行商问题进行建模,并建立二者之间的转换关系,论证了TSP可行解的充分必要条件是对应分配问题的可行解与辅助边结合后仅包含一个环路,对6个标准旅行商进行测试和对比分析,验证算法的有效性。实验结果表明:在不同的数据集中,改进匈牙利算法能有效求出TSP问题的解。

基于视觉的交通场景目标检测与跟踪技术研究

针对通用目标检测算法在交通场景下检测多目标时存在的小目标漏检和误检问题,提出了一种适用于交通场景下的目标检测与跟踪算法。基于K-means++聚类算法计算出适合交通场景下目标检测的锚定框(Anchor Box)和深度可分离卷积的思想优化目标检测算法。其次,利用匈牙利匹配将卡尔曼滤波的预测值和目标检测的实际值相关联匹配,以实现目标跟踪。实验结果证明,改进后的目标检测与跟踪算法较原始算法在mAP方面提升了2.3%,达到了99.13%;速度提升了20%,达到了30帧/秒,对交通场景下的特定目标均实现了准确检测跟踪,保持了较好的实时性。

基于需求密度预测的网约车集约化调度方法

为提升网约车接单率和利润率、实现全局供需平衡,提出一种基于需求密度预测的网约车集约化调度方法。首先,根据历史数据设计基于多层混合感知野的深度时空残差感知网络结构,该网络基于需求频度划分历史时空数据,并通过卷积指数线性网络及残差单元对不同时空数据进行差异化处理。结合基于门控机制的融合及求和融合方法动态聚合时间、空间和外部特征,实现了对需求密度的准确预测,从而预估网约车需求密度集群效益。其次,基于网约车经济效益和需求密度集群效益,建立调度数学模型,设计传感邻域限制调度范围,提升搜索效率。将遗传算法与匈牙利算法相结合,提高算法寻优能力,避免基因缺失,通过改进选择和变异算子,增强遗传算法的局部随机搜索能力,规避早熟风险,从而得到网约车与乘客的最佳匹配结果,保证了全局供需平衡和总体盈利能力。最后,基于大规模真实数据集对预测模型的性能和调度算法的有效性进行验证,实验结果表明,预测模型精度可达到97%,调度算法的求解质量可达最优解的99%,可为网约车平台提供调度策略,保障交通系统稳定。

基于TSI的匈塞铁路塞段站场主要技术标准研究

研究目的:为指导匈塞铁路塞尔维亚段站场工程设计、施工,并便于建成后的运输组织和养护维修,结合项目所在国铁路运输、设备维护及管理需要,基于欧盟铁路互联互通(TSI)条件,对线间距、到发线有效长度、客运设备及车辆设备、站线轨道四方面技术标准展开对比研究。研究结论:(1)车站线间距的规定推荐采用塞尔维亚标准;(2)接发客货列车的到发线有效长度一般为650 m,每隔20~25 km车站到发线有效长度满足750 m;仅接发客车的到发线有效长度:临靠400 m长度站台的到发线,有效长度不小于530 m,临靠220 m长度站台的到发线,有效长度不小于350 m;(3)停靠国际列车的站台长度不少于400 m,仅停靠国内列车的站台长度不少于220 m,侧式站台宽度为4 m,岛式站台宽度为6.66 m,站台高度采用0.55 m;(4)正线和到发线采用60E1型,R260品质钢轨;其余站线轨道采用49E1型,R260品质钢轨;道岔选用UIC道岔系列,根据通过速度选择不同型号;(5)本研究成果对其他欧洲铁路项目站场主要技术标准的选择及国内铁路设计具有借鉴意义。

基于红外图像处理的移动目标识别跟踪算法

为解决红外热像仪在战场环境多变、背景复杂、目标运动速度过快等情况下难以对目标进行识别与跟踪的问题,本文提出了一种基于红外图像处理的移动目标识别跟踪算法。该算法主要分为移动目标的检测和跟踪两部分。首先采用高斯背景建模方法来提取运动目标,其次应用Kalman滤波来进行运动目标的定位与跟踪,最后通过匈牙利算法来判定目标点的归属。实验结果表明,在MOT基准挑战中的评价指标上,本文算法相比于Sort跟踪器将MOTA提升了1.4%,MT提升了11.9%,ML降低了6.5%,IDs降低了33.3%,FN降低了13.6%,提高了跟踪准确性,成功实现了对多个移动目标的跟踪。本文算法在军事领域具有广泛的应用前景。

“西方马克思主义者”与“匈牙利知识分子”:卢卡奇身份问题之考察

卢卡奇是20世纪西方最为重要的思想家之一,他的作品在20世纪30年代就被翻译成了中文,对国内现实主义文论发展起到了一定作用。但受限于苏联学界的影响,学界一直未能深入了解、发掘卢卡奇理论及其价值。改革开放后,国内对卢卡奇的研究热度逐渐上升并主要关注其哲学问题。人们往往更多地关注卢卡奇作为“西方马克思主义者”的身份,而忽视了其匈牙利知识分子的身份。将卢卡奇的经历与思想,重新置于西方马克思主义形成与匈牙利社会思想史中,方能正确认识卢卡奇的身份问题。

基于加权窗口特征的多目标火力集中分配和动态调整算法

针对现有的火力打击规划方法在战前筹划时存在局部收敛性和全局优化能力差、缺乏对战中调整的临机处理和时效性考虑等问题,提出了基于加权窗口特征的火力集中分配和动态调整算法。首先,建立了基于加权窗口特征的分配模型以量化打击效益和资源代价,从而适应对不同作战需求的火力分配;然后,针对算法参数敏感和先验知识缺失的问题,提出了一种自适应参数的基于密度的聚类算法,可自适应寻优参数选择;最后,考虑到战中调整的实时性需求,提出了改进匈牙利算法以减少代价矩阵的迭代次数。试验表明,该算法能够有效解决战前规划和战中调整问题,且在大规模数据集上表现出比其他方法更加优越的性能,验证了该算法的有效性和灵活性。

车联网中基于进化策略算法与匈牙利算法的资源分配策略

为了提高车联网中高清地图下载业务的吞吐量和降低车队行驶业务的传输时延,提出一种基于进化策略算法和匈牙利算法(Evolutionary Strategy Algorithm and Hungarian Algorithm,ES-HA)的网络切片资源分配策略。构建增强型移动带宽(Enhanced Mobile Broadband,eMBB)切片和高可靠低时延(Ultra Reliable&Low Latency Communication,uRLLC)切片,根据eMBB用户和uRLLC用户功率之间的函数关系求得最佳功率,采用ES算法获得两种用户的最佳带宽,并使用HA实现最佳信道匹配。仿真结果表明,与基于集群的资源块共享和功率分配(Cluster-based Resource Block Sharing and Power Allocation,CROWN)算法、基于基准算法的资源分配策略在总吞吐量、传输任务时延、链路容量及最小吞吐量方面进行对比,该策略在满足车到基础设施(Vehicle to Infrastructure,V2I)链路用户高容量需求的同时,能够提高下载业务的吞吐量和降低车队行驶业务的传输时延。

改进关联策略的三维多目标跟踪算法

为提高基于激光雷达的三维多目标跟踪准确度,提出一种基于检测的3D多目标跟踪算法。使用深度神经网络从激光点云中获取目标的位置信息后,对目标跟踪算法中的关联策略进行了优化。首先,估计出目标的速度信息,与位置信息一同纳入观测值,在BEV视角下使用卡尔曼滤波器对目标的状态进行预测与更新;然后,基于目标与激光雷达的距离来评估目标位置的不确定度,用于修正观测模型中的协方差矩阵;最后,在马氏距离中添加对目标测量的不确定性加权项,使用匈牙利算法进行数据关联。在大规模自动驾驶数据集Nuscenes上对所提算法进行了测试,得出其跟踪精度超过了现有的基线方法。消融实验结果表明,所提出的改进措施能有效提高三维多目标跟踪的性能。

单机流式货架系统中双命令操作的生成方法

单机流式货架系统作为经典流式货架系统的一种,具有高存储密度、高空间利用率等特点。其存储单元被设计成狭窄的U型,使得入库端和出库端位于同一个工作面彼此相邻的位置。存取机可以采用单命令模式和双命令模式执行入库作业和出库作业。文章根据匈牙利算法,为单机流式货架系统生成双命令操作序列,并通过仿真实验验证匈牙利算法生成双命令操作的有效性。

Study on the Hungarian algorithm for the maximum likelihood data association problem

A specialized Hungarian algorithm was developed here for the maximum likelihood data association problem with two implementation versions due to presence of false alarms and missed detections. The maximum likelihood data association problem is formulated as a bipartite weighted matching problem. Its duality and the optimality conditions are given. The Hungarian algorithm with its computational steps, data structure and computational complexity is presented. The two implementation versions, Hungarian forest （HF） algorithm and Hungarian tree （HT） algorithm, and their combination with the naYve auction initialization are discussed. The computational results show that HT algorithm is slightly faster than HF algorithm and they are both superior to the classic Munkres algorithm.

Improved Hungarian algorithm for assignment problems of serial-parallel systems

In order to overcome the shortcoming of the classical Hungarian algorithm that it can only solve the problems where the total cost is the sum of that of each job, an improved Hungarian algorithm is proposed and used to solve the assignment problem of serial-parallel systems. First of all, by replacing parallel jobs with virtual jobs, the proposed algorithm converts the serial-parallel system into a pure serial system, where the classical Hungarian algorithm can be used to generate a temporal assignment plan via optimization. Afterwards, the assignment plan is validated by checking whether the virtual jobs can be realized by real jobs through local searching. If the assignment plan is not valid, the converted system will be adapted by adjusting the parameters of virtual jobs, and then be optimized again. Through iterative searching, the valid optimal assignment plan can eventually be obtained.To evaluate the proposed algorithm, the valid optimal assignment plan is applied to labor allocation of a manufacturing system which is a typical serial-parallel system.

Competition assignment problem algorithm based on Hungarian method

Traditional Hungarian method can only solve standard assignment problems, while can not solve competition assignment problems. This article emphatically discussed the difference between standard assignment problems and competition assignment problems. The kinds of competition assignment problem algorithms based on Hungarian method and the solutions of them were studied.

Hungarian算法的推广

分析了非平衡分派问题的线性规划模型的性质和特点，在此基础上，将Hugarian算法推广到非平衡分派问题，利用该算法，可以直接求出非平衡分派问题．

Multi-Objective Optimization of Time-Cost-Quality Using Hungarian Algorithm

In this paper, we propose an algorithm for solving multi-objective assignment problem (MOAP) through Hungarian Algorithm, and this approach emphasizes on optimal solution of each objective function by minimizing the resource. To illustrate the algorithm a numerical example (Sec. 4;Table 1) is presented.

Adaptive subcarrier allocation for MIMO-OFDMA wireless systems using Hungarian method

MIMO-OFDMA is a promising technique for future broadband wireless communication systems. In this paper, the problem of allocating subcarriers among different users to maximize the total capacity is addressed. The optimal solution can be obtained by Hungarian method is proved, using two utility matrices, i.e. the Frobenius-norm matrix and the determinant matrix. Simulation results show that the proposed algorithm can achieve higher capacity than the existing algorithms, and it is globally optimal and easy to be implemented.

An Application of the Hungarian Algorithm to Solve Traveling Salesman Problem

In this paper we applicate the Hungarian algorithm for assignment problem to solve traveling salesman problem. Tree examples of application of algorithm are included.

Tinnitus characteristics and associated variables on Tinnitus Handicap Inventory among a Hungarian population

Background:Tinnitus is a sound precepted without an external sound stimulus.Its background can be categorised into primary and secondary cases.The secondary cases include pathologies of the external,middle and inner ear.Tinnitus can be objective or subjective;the latter can only identified by the sufferer.Previous research results have shown that tinnitus significantly affects the quality of life and daily functioning.Objectives:To analyse the impact of tinnitus on the daily functioning and the possible influence of demographical data and tinnitus duration on it.Methods:630 patients(265 males and 365 females,25e85 years of age)suffering from primary tinnitus were enrolled.In the Hungarian language,these patients completed the Tinnitus Handicap Inventory(THI)questionnaire and underwent a complete otorhinolaryngological examination.IBM SPSS V24 software was used for data processing;correlation tests,the Mann-Whitney U and Kruskal-Wallis nonparametric tests were used.Results:According to the THI questionnaires outcomes,most patients(62.5%)were presented with a mild handicap.Based on statistical analysis,no significant correlation was observed between the total THI points and the age of the patients,along with the duration and localisation of the symptoms.However,the total THI scores of male and female patients significantly differed,indicating higher THI values in the female group(p=0.00052*).Conclusions:The tinnitus severity was not affected by the duration,localisation of the symptoms and age but by gender,indicating higher values in the case of females.