Rail-Pillar Net:A 3D Detection Network for Railway Foreign Object Based on LiDAR

Aiming at the limitations of the existing railway foreign object detection methods based on two-dimensional(2D)images,such as short detection distance,strong influence of environment and lack of distance information,we propose Rail-PillarNet,a three-dimensional(3D)LIDAR(Light Detection and Ranging)railway foreign object detection method based on the improvement of PointPillars.Firstly,the parallel attention pillar encoder(PAPE)is designed to fully extract the features of the pillars and alleviate the problem of local fine-grained information loss in PointPillars pillars encoder.Secondly,a fine backbone network is designed to improve the feature extraction capability of the network by combining the coding characteristics of LIDAR point cloud feature and residual structure.Finally,the initial weight parameters of the model were optimised by the transfer learning training method to further improve accuracy.The experimental results on the OSDaR23 dataset show that the average accuracy of Rail-PillarNet reaches 58.51%,which is higher than most mainstream models,and the number of parameters is 5.49 M.Compared with PointPillars,the accuracy of each target is improved by 10.94%,3.53%,16.96%and 19.90%,respectively,and the number of parameters only increases by 0.64M,which achieves a balance between the number of parameters and accuracy.

Development of vehicle-recognition method on water surfaces using LiDAR data:SPD^(2)(spherically stratified point projection with diameter and distance)

Swarm robot systems are an important application of autonomous unmanned surface vehicles on water surfaces.For monitoring natural environments and conducting security activities within a certain range using a surface vehicle,the swarm robot system is more efficient than the operation of a single object as the former can reduce cost and save time.It is necessary to detect adjacent surface obstacles robustly to operate a cluster of unmanned surface vehicles.For this purpose,a LiDAR(light detection and ranging)sensor is used as it can simultaneously obtain 3D information for all directions,relatively robustly and accurately,irrespective of the surrounding environmental conditions.Although the GPS(global-positioning-system)error range exists,obtaining measurements of the surface-vessel position can still ensure stability during platoon maneuvering.In this study,a three-layer convolutional neural network is applied to classify types of surface vehicles.The aim of this approach is to redefine the sparse 3D point cloud data as 2D image data with a connotative meaning and subsequently utilize this transformed data for object classification purposes.Hence,we have proposed a descriptor that converts the 3D point cloud data into 2D image data.To use this descriptor effectively,it is necessary to perform a clustering operation that separates the point clouds for each object.We developed voxel-based clustering for the point cloud clustering.Furthermore,using the descriptor,3D point cloud data can be converted into a 2D feature image,and the converted 2D image is provided as an input value to the network.We intend to verify the validity of the proposed 3D point cloud feature descriptor by using experimental data in the simulator.Furthermore,we explore the feasibility of real-time object classification within this framework.

Multi-Scale Feature Extraction for Joint Classification of Hyperspectral and LiDAR Data

With the development of sensors,the application of multi-source remote sensing data has been widely concerned.Since hyperspectral image(HSI)contains rich spectral information while light detection and ranging(LiDAR)data contains elevation information,joint use of them for ground object classification can yield positive results,especially by building deep networks.Fortu-nately,multi-scale deep networks allow to expand the receptive fields of convolution without causing the computational and training problems associated with simply adding more network layers.In this work,a multi-scale feature fusion network is proposed for the joint classification of HSI and LiDAR data.First,we design a multi-scale spatial feature extraction module with cross-channel connections,by which spatial information of HSI data and elevation information of LiDAR data are extracted and fused.In addition,a multi-scale spectral feature extraction module is employed to extract the multi-scale spectral features of HSI data.Finally,joint multi-scale features are obtained by weighting and concatenation operations and then fed into the classifier.To verify the effective-ness of the proposed network,experiments are carried out on the MUUFL Gulfport and Trento datasets.The experimental results demonstrate that the classification performance of the proposed method is superior to that of other state-of-the-art methods.

林区机载LiDAR点云的多分辨率层次布料模拟滤波

针对现有机载LiDAR(light detection and ranging)点云滤波方法在地形起伏剧烈的林区适用性不足的问题,提出一种多分辨率层次布料模拟滤波方法。首先,通过多尺度形态学开运算选择大量种子地面点;然后,基于种子地面点,使用布料模拟法由低至高逐层构建参考地形,以快速获取高分辨率参考地形;最后,基于点至参考地形的高差区分地面点和非地面点。利用国际摄影测量和遥感学会提供的数据集和参考方法,评估该方法性能。利用在中国、美国多个代表性林区的点云数据,评估该方法的可推广性。结果表明,该方法的Kappa系数和运行时间是83.72%和34.11 s,精度和效率较经典布料模拟滤波方法提高10.49%和52.17%。相比8种参考方法,该方法能够获得更高精度,并且具有稳定的可推广性。

Three-channel CMOS transimpedance amplifier for LiDAR sensor receiver

For time-of-flight(TOF)light detection and ranging(LiDAR),a three-channel high-performance transimpedance amplifier(TIA)with high immunity to input load capacitance is presented.A regulated cascade(RGC)as the input stage is at the core of the complementary metal oxide semiconductor(CMOS)circuit chip,giving it more immunity to input photodiode detectors.A simple smart output interface acting as a feedback structure,which is rarely found in other designs,reduces the chip size and power consumption simultaneously.The circuit is designed using a 0.5μm CMOS process technology to achieve low cost.The device delivers a 33.87 dB?transimpedance gain at 350 MHz.With a higher input load capacitance,it shows a-3 dB bandwidth of 461 MHz,indicating a better detector tolerance at the front end of the system.Under a 3.3 V supply voltage,the device consumes 5.2 mW,and the total chip area with three channels is 402.8×597.0μm2(including the test pads).

基于无人机载LiDAR点云数据的电力线提取方法

机载激光雷达(LiDAR)点云电力线提取过程中存在杆塔形状复杂、噪声影响大等问题,导致电力线点云提取精度低,本文提出一种基于点云分块处理、格网划分的曲面拟合滤波、自适应密度聚类算法的电力线点云提取与重建方法。首先,根据电力线走向,对整体点云进行分块处理;其次,在曲面拟合算法的基础上,引入格网划分思想,提出一种改进曲面拟合滤波算法并进行点云滤波;最后,通过给出自适应密度聚类解决方案精确提取电力线点云。借助点云库(PCL)、libLAS库与Visual Studio 2017 C++开发环境实现本文算法,基于实测点云数据对本文方法进行测试与精度评定。结果表明:电力线提取精确率为97.82%、召回率为99.76%、F1值为98.78%,一次便可实现电力线的成功提取,在保证提取精度的同时提升了提取效率,本文研究能够为电力线智能巡检提供良好的工程应用价值。

无人机LiDAR免控制测量技术在1∶10000 DEM更新中的应用

在安徽省1∶10000基础地理信息数据更新工程DEM局部更新工作中,针对外业中控制点布设工作量大、难度高的特点,研究无人机LiDAR测量技术生产DEM的免控制技术方法,利用激光点云分类结果中提取的可靠地面点,建立高程拟合模型,拟合转换点云高程,生产DEM。并与传统通过实测控制点进行点云高程拟合生产的DEM成果精度比较。研究结果表明:利用免控制技术方法生产的DEM精度可以满足安徽省1∶10000 DEM更新要求。

Laser diode drive method with narrow-width and high-peak current for multi-line LIDAR

Light detection and ranging (LIDAR) based on time of flight (TOF) method is widely used in many fields related to distance measurement. LIDAR generally uses laser diode (LD) to emit the pulsed laser with high peak power and short duration to ensure a large distance measurement range and eye safety. To achieve this goal, we propose a pulsed LD drive method producing the drive current with high peak and narrow pulse width. We analyze the key issues and related theories of the drive current generation based on this method and design an LD driver. A model of drive current generation is established and the influence of operating frequency on drive current is discussed. The LD driver is simulated by software and verified by experiments. The working frequency of the driver changes from 20 kHz to 100 kHz and the charging voltage is set at 130 V. The current produced by this driver has a duration of 8.8 ns and a peak of about 35 A, and the peak output optical power of the LD exceeds 75 W.

高精度LiDAR技术在活动断层断错地貌研究中的应用——以沂沭断裂带莒县至郯城段为例

郯庐断裂带是我国东部地区最为重要的大型活动断裂之一,具有较强的发震能力,断裂带沿线多处穿过人类活动及聚集的城市。为了研究郯庐断裂带的几何展布及活动性特征,选取郯庐断裂带的沂沭断裂莒县至郯城范围为研究对象,考虑东部地区植被覆盖,采用机载激光雷达和地基激光雷达手段采集了断裂带沿线左山(一步涧)段、钟华山段、岌山段、马陵山段以及蒋家岭等地高精度地形数据,对断裂的滑动分布特征进行分析。通过对测量得到的点云数据进行点云匹配、镶嵌、植物滤除以及不规则三角网建模处理,得到了0.1 m空间分辨率的数字高程模型(Digital Elevation Model,DEM),实现了对断裂微地貌形态的高清晰度三维再现,提取了断裂断错冲沟右旋水平位错量和断层陡坎垂直位错量,结合地质填图和探槽揭露的古地震事件进行验证和对比,对郯庐断裂带莒县至郯城段的断层活动特征进行了定量分析。结果表明,沿断裂带左山(一步涧)段、钟华山段、岌山段和马陵山段的水平和垂直位错量具有分级特征,晚第四纪以来可能发生过多期(3~5次)活动和多次古地震事件。研究成果为进一步研究郯庐断裂运动学和几何学提供了重要依据,同时也说明地基激光雷达和机载激光雷达技术在活动断层研究中有着广泛的应用前景。

不同森林覆盖类型对LiDAR生成DEM的精度影响分析

运用机载激光雷达(LiDAR)获取的高精度数字高程模型(DEM)是目前地学分析应用满足精细化要求的基础数据。目前对森林覆盖下DEM产品的精度影响分析依然有所欠缺。以实时动态载波相位差分技术(RTK)结合全站仪实测的高程碎部点与机载LiDAR数据生成的高精度DEM产品数值做差,进行单因素方差分析(ANOVA)检验以及独立样本T-检验。经多重比较分析(LSD)可知,以马尾松为代表的针叶林和以桉树、茶树、黄连木、荔枝、阔叶混合林为代表的阔叶林之间差值相差较大,说明针叶林和阔叶林两种不同森林覆盖类型的绝对精度存在差异。但Robust检验结果显示阔叶林地表DEM的精度均值相等性为1,说明阔叶林(以桉树、茶树、黄连木、荔枝、阔叶混合林林地为代表)地表DEM精度不受树种的影响。综上,阔叶林与针叶林的地表DEM精度差异显著,说明阔叶林和针叶林对机载LiDAR生成森林区域地表DEM存在一定的影响。

倾斜摄影和LiDAR技术融合在大比例尺地形图测绘中的应用

地形图传统测量方法费时费力。本文融合倾斜摄影与激光探测及测距系统技术,以抚州市比亚迪厂区地形图测绘项目为例,非房屋区域采用机载激光雷达航测技术,房屋区域采用倾斜摄影测量航测技术成图。最后通过外业采集检查点与内业测图成果进行精度对比分析。结果表明:该技术测绘成果平面与高程均能满足大比例尺测图的精度要求,而且能节约地形图制作时间及成本。

基于机载LiDAR点云的建筑物多细节层次建模方法

机载激光雷达点云(LiDAR)重建大范围建筑物模型一直是实景三维建模的热点问题,由于机载激光点云稀疏、建筑物立面数据缺失,给模型自动化重建带来了极大的挑战。为了解决该问题,本论文研究基于图优化理论重建建筑物多细节层次(LOD)模型。首先,对原始点云进行自动化滤波,分离地面点和非地面点,对地面点构建数字高程模型(DEM),对非地面点半自动提取单体化建筑物点云。然后,基于滚球法(Alpha shape)提取建筑物边界,利用通用图优化方法(G2O)对误差线进行全局一致性改正,获得规则化的二维边框。并基于建筑物屋顶三维高程及DEM高程值重建建筑物LOD1模型。其次,根据点云的高程差异生成高程栅格图,从高程栅格图提取建筑物轮廓线,对轮廓线进行简化、规则化、聚类,并将规则后的边界线拉伸获取建筑物立面结构,弥补建筑物立面数据缺失对建模的影响。最后,将屋顶平面相交、建筑物立面裁剪,对候选平面进行二元图割全局优化,选择最能表达建筑物结构的平面,以此重建建筑物LOD2模型。本论文选择北京市2017年机载激光点云进行实验,结果表明,本文提出的方法可以稳健地重建建筑物多细节层次模型,LOD2模型距离偏差为0.22 m,LOD1模型距离偏差为0.71 m,整体模型精确度高;针对1147个建筑物,LOD2和LOD1模型重建时间分别为1290s和1097s,具有较高的自动化程度和建模效率。

基于机载LiDAR点云的电力线提取与三维重建

为了解决地形复杂、点云密度不均匀的输电线机载激光雷达(LiDAR)点云电力线提取精度低的问题,本文根据电力线点的空间分布特征设计与实现了一套电力线提取与三维重建方法。首先,使用改进曲面拟合滤波算法与形态学开运算实现地面点、低矮植被点等的滤除;其次,以滤波处理得到点云数据为数据源,利用电力线点维度特征实现电力线点粗提取并利用密度聚类算法进行单根电力线精提取;最后,基于单根电力线提取结果进行电力线三维重建。为了对本文提出电力线提取与重建方法进行检验,使用宁波市某高压交流输变电工程中部分实测机载LiDAR点云数据进行实验,结果表明,本文方法提取28根电力线结果误差率均在0.04%以内,验证了本文方法的可靠性与实用性。

机载LiDAR河流高程拟合算法

使用机载激光雷达(LiDAR)进行数字高程模型(DEM)制作过程中,河流数据缺失,需进行人工编辑,目前处理流程中难以既保证数据精度又保证整体平整美观,本文提出机载LiDAR河流高程拟合方案,通过半自动河流边线提取、河流中心线提取以及中心线高程拟合一系列技术流程,不仅解决当前复杂的人工编辑问题,还提升了拟合精度,文中详细地阐述河流高程拟合关键算法,并基于Micro station V8开发出应用实例,为机载LiDAR河流高程拟合提供新思路。

面向机载LiDAR点云数据的双线河水体DEM构建方法

机载激光雷达(LiDAR)系统可以快速获取高精度的地面点云数据,可以快速生成数字高程模型(DEM)。但是基于原始水体点云数据直接构建DEM效果较差,针对这个问题,论文提出了一种双线河水体点云填充方法,对缺失的河道内点云数据进行填充,并设计开发了河流优化填充模块,实现了双线河水体具有上下游特征的DEM构建效果。

多场景机载LiDAR点云单木提取

机载激光雷达(LiDAR)测量技术点云数据采集效率高、覆盖范围大,在树木三维空间采集方面具有显著的优势。本文在城市绿化林地、农村果树林地、山区经济林地这三种典型场景中各选取了一块实验区,采用多级分类的思路,通过粗差点检测、点云滤波、高度阈值分割、平面生长等处理步骤,对整个区域的点云数据进行语义标号,相关的语义类别包括:粗差、地面、植被和建筑物,再使用谱聚类算法对植被类点云进行单木提取。实验结果表明,城市绿化林地树木分布规整,树冠彼此相对独立,检测率最高(83.6%);而农村果树林地存在植被点与房屋点更难区分,低矮树木漏识别多于城市场景,检测率略低于城市场景(81.2%);山区经济林地树木密度最高,树冠相互粘连、遮盖,检测率最低(76.8%)。

激光雷达超远距离测距技术

针对超远距离多功能交会对接激光雷达需求,开展基于非相干测距技术的远距离激光测距通信一体化模块研制,在不改变原有雷达主机架构和信号体制下,实现对远距离高动态合作目标的通信测距功能。推导出测距原理,对动态、时钟性能等因素产生的测距误差进行理论分析,给出速度、时钟性能对测距误差的影响公式。得出在高动态环境下,相对速度与测距周期、双方钟差共同作用产生测距系统误差,且速度越大系统误差越大的结论。设计测距通信一体化演示验证平台,完成测距通信算法的软硬件评估,实测结果与理论推导相符,为后续新体制激光雷达原理样机研制奠定技术基础。

城市森林结构多样性预测冠下地面温度的潜力研究

城市森林冠层具有调控城市森林微气候的能力,但现有研究尚未阐明冠层结构对冠下地面温度的影响及其预测潜力。文章基于无人机机载激光雷达(UAV-LiDAR)提取哈尔滨林业示范基地的城市森林冠层结构多样性特征指标,探究单一结构多样性特征对冠下地面温度的影响,以及结构多样性多因子组合对温度的预测潜力。结果表明:1)城市森林结构多样性的8个特征因子与冠下地面温度呈显著相关关系(P<0.05),其中深间隙(DG)、深间隙分数(DGF)、覆盖分数(CF)、间隙分数分布(GFP)表征了结构多样性的覆盖/开放度特征;冠层高度标准差(H_(std))、冠层高度最大值(H_(max))、95%分位点高度(ZQ_(95))表征了高度特征;垂直复杂指数(VCI)表征了异质性特征。2)城市森林冠层结构多样性的覆盖/开放度特征对冠下地面温度的响应更强(R^(2)为0.15~0.5),强于高度指标(R^(2)为0.14~0.19)以及异质性指标(R^(2)=0.14)。3)结合高度指标、覆盖/开放度指标以及异质性指标的多因子预测模型2(R^(2)=0.61,RMSE=0.51,MSE=0.26,AIC=62.74),对于冠下地面温度的预测性能更优。研究明晰了城市森林结构多样性的多因子变量及其特征组合预测冠下地面温度的潜力,为城市森林冠层结构调控内部小气候环境研究提供了科学参考。

激光雷达与相机联合标定进展研究

针对激光雷达与相机标定精度会极大地影响定位初始化,融合定位算法对标定参数的准确性非常敏感,会严重降低融合定位算法的性能和可靠性等问题,提出一种激光雷达与相机联合标定分类方法:论述激光雷达与相机联合标定领域的最新的研究进展;重点分析和总结基于特征的标定方法、基于运动的标定方法、基于互信息的标定方法和基于深度学习的标定方法的突出研究成果;总结出4种标定方法的开源代码工具集,并对这4种标定方法的特点、标定精度和自动化程度进行分析比较;最后,展望激光雷达与相机联合标定研究的发展趋势。

集成多源遥感与极限梯度提升的竹林地上生物量估测

森林生物量遥感精准估测是评估森林碳汇潜力的重要依据,通过对比机载激光雷达(LiDAR)与机载多光谱数据在竹林地上生物量(AGB)估测中的差异,探索不同光学与激光点云数据协同估测森林AGB的潜力。以福建特色竹种黄甜竹(Acidosasa edulis)人工林为研究对象,同步采集无人机载多光谱影像和LiDAR数据,提取并优选单一LiDAR的优选特征集、单一多光谱影像数据优选特征集、联合机载LiDAR和多光谱影像的优选特征集,采用多元线性回归模型(MLR)和支持向量机(SVM)、随机森林(RF)、极限梯度提升(XGBoost)拟合竹林AGB估测模型。结果表明:(1)LiDAR数据的高度特征和密度特征是竹林AGB估测的重要特征,基于多源异构数据的特征集对竹林AGB估测的效果优于单一的遥感数据;(2)竹林AGB估测中,基于机器学习算法的非参数模型优于多元线性回归模型;集成XGBoost与多源特征集的黄甜竹AGB估测模型最优,R^(2)达0.64,Erms为9.90 t·hm^(-2)。竹林具有特殊冠层结构,应用LiDAR数据源能够有效获取林分垂直结构信息,联合多源异构数据能提高生物量估测模型精度。