Machine Learning-based Identification of Contaminated Images in Light Curve Data Preprocessing

Attitude is one of the crucial parameters for space objects and plays a vital role in collision prediction and debris removal.Analyzing light curves to determine attitude is the most commonly used method.In photometric observations,outliers may exist in the obtained light curves due to various reasons.Therefore,preprocessing is required to remove these outliers to obtain high quality light curves.Through statistical analysis,the reasons leading to outliers can be categorized into two main types:first,the brightness of the object significantly increases due to the passage of a star nearby,referred to as“stellar contamination,”and second,the brightness markedly decreases due to cloudy cover,referred to as“cloudy contamination.”The traditional approach of manually inspecting images for contamination is time-consuming and labor-intensive.However,we propose the utilization of machine learning methods as a substitute.Convolutional Neural Networks and SVMs are employed to identify cases of stellar contamination and cloudy contamination,achieving F1 scores of 1.00 and 0.98 on a test set,respectively.We also explore other machine learning methods such as ResNet-18 and Light Gradient Boosting Machine,then conduct comparative analyses of the results.

基于NPP-VIIRS的胶东经济圈GDP时空演变及空间化模拟研究

为研究胶东经济圈经济发展的空间差异,以2012-2022年NPP-VIIRS夜间灯光数据和GDP数据为数据源,运用探索性空间分析方法探索GDP时空演变特征,采用指数平滑、空间建模和逐像元线性矫正等方法进行GDP空间化模拟和预测.结果表明:1)胶东经济圈最佳夜间灯光指数和GDP高度相关,有相似的空间分布和演变规律;2)经济欠活跃型区域逐步缩减,经济一般型区域空间分布由带状逐渐转为片状,GDP的空间聚集性趋于减弱;3)GDP热点区主要在青岛的黄岛区和胶州市,冷点区主要在潍坊的部分县区且逐渐收缩;4)指数平滑后的夜间灯光数据与GDP相关性显著增强,更适合进行拟合;5)GDP空间化模拟相对误差小于1%,模拟效果较好;6)高密度GDP主要分布在沿海县区和城市中心城区,并在各县区的交通线节点上零星出现.

基于时序NPP-VIIRS夜间灯光和土地利用数据的福州市GDP时空分布研究

基于时序NPP-VIIRS夜间灯光、社会经济统计等数据,构建GDP空间化模型,并结合空间分辨率为30 m×30 m的土地利用数据对模型进行修正,研究2013—2020年福州市经济空间差异性及演变趋势。结果表明:(1)利用土地利用数据对GDP空间化模型进行修正以提高单元网格精度和效度的方法是可行的,对较小尺度的经济空间分析具有意义。(2)将土地利用数据对应的各产业产值用于对GDP空间化模型修正,能够进一步提升单元网格的模拟精度。(3)从共时性比对来看,福州市的经济空间呈现东部GDP较高,西部和北部GDP以低值为主、高值零星分布的格局。从历时性来看,自2015年起,福州市GDP高值区域的扩张与GDP总值的增长速度明显提高,2018年后整体出现略微下降趋势。

Spatiotemporal Evolution of West Africa's Urban Landscape Characteristics Applying Harmonized DMSP-OLS and NPP-VIIRS Nighttime Light(NTL)Data

Investigating urban expansion patterns aids in the management of urbanization and in ameliorating the socioeconomic and environmental issues associated with economic transformation and sustainable development.Applying Harmonized Defense Meteorological Satellite Program-Operational Line-scan System(DMSP-OLS)and the Suomi National Polar-Orbiting Partnership-Visible Infrared Imagery Radiometer Suite(NPP-VIIRS)Nighttime Light(NTL)data,this paper investigated the characteristics of urban landscape in West Africa.Using the harmonized NTL data,spatial comparison and empirical threshold methods were employed to detect urban changes from 1993 to 2018.We examined the rate of urban change and calculated the direction of the urban expansion of West Africa using the center-of-gravity method for urban areas.In addition,we used the landscape expansion index method to assess the processes and stages of urban growth in West Africa.The accuracy of urban area extraction based on NTL data were R^(2)=0.8314 in 2000,R^(2)=0.8809 in 2006,R^(2)=0.9051 in 2012 for the DMSP-OLS and the simulated NPP-VIIRS was R^(2)=0.8426 in 2018,by using Google Earth images as validation.The results indicated that there was a high rate and acceleration of urban landscapes in West Africa,with rates of 0.0160,0.0173,0.0189,and 0.0686,and accelerations of 0.31,0.42,0.54,and 0.90 for the periods of 1998–2003,2003–2008,2008–2013,and 2013–2018,respectively.The expansion direction of urban agglomeration in West Africa during 1993–2018 was mainly from the coast to inland.However,cities located in the Sahel Region of Africa and in the middle zone expanded from north to south.Finally,the results showed that the urban landscape of West Africa was mainly in a scattered and disordered’diffusion’process,whereas only a few cities located in coastal areas experiencing the process of’coalescence’according to urban growth phase theory.This study provides urban planners with relevant insights for the urban expansion characteristics of West Africa.

DMSP-OLS与NPP-VIIRS夜间灯光遥感影像数据整合

DMSP-OLS夜间灯光遥感数据截至2013年,现已被NPP-VIIRS夜间灯光数据取代。因此,要获得长时间序列且稳定的夜间灯光数据集,需要整合两类夜间灯光数据。基于此,本文提出了基于重采样的两类数据整合方法,对2013—2020年NPP-VIIRS数据进行模拟,最终建立了1992—2020年长时间序列校正—模拟DMSP-OLS夜光遥感数据集。结果表明,基于重采样的整合方法效果良好(城市区域Pearson相关系数ρ=0.985 2,RMSE=3.460 7),整合数据集与相关社会经济参考量高度契合(影像DN值总和与GDP的相关系数ρ=0.946,与人口的相关系数ρ=0.971,二次多项式模型拟合R~2≈0.98,RMSE<5.55),优于已有研究。因此,利用该方法整合后的数据集能更好地支撑基于夜间灯光影像的长时间序列研究。

基于NPP-VIIRS夜间灯光数据的福建省县域经济发展时空演化研究

基于福建省县域统计经济数据和校准预处理后的NPP-VIIRS夜间灯光数据,揭示县域统计经济数据与夜间灯光数据的关联,在此基础上探讨县域经济发展时空演化特征。结果表明:(1)夜间灯光总量与地区生产总值(GDP)高度相关,县域夜间灯光总量集中分布在东南沿海地区,沿海县(市)单位面积夜间灯光总量高于内陆县(市),夜间灯光总量空间分布呈现高度不均衡性;(2)县域GDP全局自相关指数以及夜间灯光全局自相关指数呈显著正相关水平,县域夜间灯光数据空间集聚水平低于县域GDP空间集聚水平,夜间灯光数据高高聚类区域相比县域GDP高高聚类区域呈扩张趋势;(3)夜间灯光数据和经济指数热点区域均分布在东南部沿海地区,冷点区域主要分布在三明和南平等地区部分县(市),夜间灯光数据和统计经济数据冷热点之间存在一致性。为缩小区域发展差异,需加大经济政策扶持,促进区域经济交流,带动经济同步增长,研究结果可为县域经济发展政策制定提供参考借鉴。

Obtaining 2D Soil Resistance Profiles from the Integration of Electrical Resistivity Data and Standard Penetration Test (SPT) and Light Dynamic Penetrometer (DPL) Resistance Tests—Applications in Mass Movements Studies

In Brazil and various regions globally, the initiation of landslides is frequently associated with rainfall;yet the spatial arrangement of geological structures and stratification considerably influences landslide occurrences. The multifaceted nature of these influences makes the surveillance of mass movements a highly intricate task, requiring an understanding of numerous interdependent variables. Recent years have seen an emergence in scholarly research aimed at integrating geophysical and geotechnical methodologies. The conjoint examination of geophysical and geotechnical data offers an enhanced perspective into subsurface structures. Within this work, a methodology is proposed for the synchronous analysis of electrical resistivity geophysical data and geotechnical data, specifically those extracted from the Light Dynamic Penetrometer (DPL) and Standard Penetration Test (SPT). This study involved a linear fitting process to correlate resistivity with N10/SPT N-values from DPL/SPT soundings, culminating in a 2D profile of N10/SPT N-values predicated on electrical profiles. The findings of this research furnish invaluable insights into slope stability by allowing for a two-dimensional representation of penetration resistance properties. Through the synthesis of geophysical and geotechnical data, this project aims to augment the comprehension of subsurface conditions, with potential implications for refining landslide risk evaluations. This endeavor offers insight into the formulation of more effective and precise slope management protocols and disaster prevention strategies.

基于NPP-VIIRS夜间灯光数据的广西西江流域城市扩张时空格局演变研究

基于2012年、2014年、2016年、2018年、2020年的年度NPP-VIIRS夜间灯光数据,运用统计数据比较法提取广西西江流域城市建成区,并计算城市建成区的扩张速度、扩张强度指数、平均灯光强度指数、紧凑度指数、城市中心迁移距离、角度和速度,分析2012—2020年广西西江流域城市扩张动态变化情况。结果表明:1)统计数据比较法的平均总体分类精度在97%以上,能精确反映西江流域城市建成区发展的实际情况。2)2012—2020年广西西江流域城市建成区平均扩张速度为4.61 km^(2)/a,呈现先快速发展后减缓的扩张趋势。扩张强度与平均灯光强度的变化趋势与扩张速度基本保持一致。3)广西西江流域城市建成区整体的紧凑度和分形维数均呈现降低的趋势,说明广西西江流域城市扩张较为平衡,存在内部建设和外部扩张两种扩张发展形式。4)2012—2020年河池市与南宁市因城市发展建设使城市中心迁移距离较大,其余各市的城市中心迁移距离平缓,稳定扩张;各城市中心迁移速度与城市中心迁移距离和方向的发展情况基本保持一致。

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.

基于注意力机制和LSTM-LightGBM的特高压直流输电线路可听噪声无效数据清洗方法

特高压直流输电线路可听噪声试验过程中,外界环境的突发性干扰会使实验数据中掺杂较多的无效数据,严重影响后续的数据分析。提出了一种基于注意力机制(attention mechanism,AM)和长短时记忆网络-轻量级梯度提升机(long short-term memory network-light gradient boosting machine,LSTM-LightGBM)的输电线路可听噪声无效数据清洗方法。首先,针对可听噪声数据的非线性、高维时序冗余特征等特点,以LSTM神经网络为基础进行特征提取;同时,引入特征维度注意力机制,自适应地分配权重来刻画关键特征信息的表达能力;进而,利用LightGBM对提取到的特征进行分类,检测出无效数据;然后,以某特高压直流输电线路实测可听噪声数据试验分析,结果表明该方法的检测精准率为95.55%,召回率为97.73%,F1分数为0.9663,均优于对比实验模型;最后,将无效数据删除并使用均值插补法填补,无效数据清洗后数据的50%值和95%值基本不变,仅降低无效数据的最大值和5%值。该算法对提高输电线路可听噪声数据的可靠性具有一定参考意义。

基于改进LightGBM的电力通信数据流量异常检测方法

针对电力通信网络攻击模式多变、检测泛化能力不足的问题,提出基于改进轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)的电力通信数据流量异常检测方法。结合最近邻规则(Edited Nearest Neighbor,ENN)算法、小波包分解技术和信息熵分析方法,提取电力通信数据流量异常特征,引入直方图算法和带深度限制的Leaf-wise生长策略,通过改进LightGBM算法建立电力通信数据流量异常检测模型,在模型中找到最优超参数配置,提高电力通信数据流量异常检测的准确率和效率。实验结果表明,设计方法在准确识别不同网络攻击类型和增强检测泛化能力方面具有显著优势,能够更好地应对电力通信网络中复杂多变的网络威胁,为电力通信系统的安全稳定运行提供有力保障。

中国地级市GDP数据的真实性:基于NPP-VIIRS夜间灯光数据的检验

文章通过使用比DMSP-OLS灯光数据更为准确的NPP-VIIRS夜间灯光数据,从更客观的角度来分析中国地级市GDP数据的真实性。首先,从主观层面对华东地区城市灯光数据和GDP数据的散点图、1995年和2013年的灯光数据对比图进行分析,得出GDP和灯光数据之间存在高度的相关性。其次以2013-2015年全国291个地级市的面板数据为样本进行固定效应分析,得出与主观层面分析一样的结果。研究结果表明灯光数据可以作为GDP的代理变量,进而可用于研究分析GDP数据是否存在水分。通过分析各地级市2013—2015年灯光数据的增长率平均值与GDP增长率平均值差值,得出东、中、西的差值平均值依次0.56%、0.62%、0.72%。

基于类NPP-VIIRS夜间灯光数据的粤港澳大湾区城市建成区时空动态特征

随着人口增长与经济发展,快速准确地获取城市建成区扩张信息,并对其时空特征进行分析,对于今后城市可持续发展具有重要意义。基于2000~2018年类NPP-VIIRS夜间灯光数据的大尺度不透水面指数提取粤港澳大湾区城市不透水面,并基于统计数据比较法得到城市建成区,对粤港澳大湾区以及各城市的建成区扩张进行时空分析。结果表明:①类NPP-VIIRS夜间灯光数据实现了对城市建成区的高精度提取,可为城市长时间序列的时空变化监测提供新的数据基础;②粤港澳大湾区城市建成区由2000年的1279.00 km^(2)增加到2018年的4730.25 km^(2),广州、深圳、东莞的建成区扩张最为明显,分别增加了793.50、812.75、1027.50 km^(2);在空间上,粤港澳大湾区城市建成区扩张主要发生在各城市之间的交界地带,逐步形成了佛山—广州—东莞—深圳—香港大型城市群核心建成区,西部地区发展相对较慢;③粤港澳大湾区城市建成区重心2000~2018年由东莞转移到广州。总之,粤港澳大湾区城市建成区在2000~2018年扩张整体呈现由快变慢、东快西慢的特点。

基于Sentinel-2A与NPP-VIIRS夜间灯光数据的城市建成区提取

利用夜间灯光数据(nighttime light data,NTL)与光学遥感影像提取城市建成区是当今的一个研究热点,其中基于植被校正的城市夜间灯光指数(vegetation adjusted NTL urban index,VANUI)被学者广泛利用,但它容易混淆城市边缘的建筑、水体,空间分辨率较低。对VANUI做出改进,提出基于建筑校正的城市夜间灯光指数(building adjusted NTL urban index,BANUI)。利用该指数对包头市南部的城市建成区进行提取,首先,借助Sentinel-2A遥感影像数据提取研究区的归一化建筑指数;然后,将其与NTL数据结合得到BANUI(空间分辨率为20 m),并由此得到空间分辨率更高、建筑信息更丰富的BANUI图像;最后,利用分水岭分割算法从BANUI,VANUI和NTL中提取出城市建成区并进行对比。结果表明,由BANUI提取的城市建成区总体精度可达93.61%,Kappa系数为0.7934,用户精度为81.34%,生产者精度为85.34%,提取结果与实际城市建成区的分布较吻合、提取精度较高,且优于另外2种数据。此方法可为NTL在城市建成区提取的研究中提供参考意见,也可用于对城市规划发展的监测。

基于NPP-VIIRS夜间灯光数据的淮河流域区域经济差异时空特征分析

基于2012-2018年校正后稳定的NPP-VIIRS夜间灯光数据,以淮河流域38市为研究对象,采用变异系数、泰尔指数和相对发展率等方法,对区域经济差异的时空特征进行分析。结果表明:淮河流域区域经济发展成效显著,但区域差异明显,整体上形成了以省域中心城市为核心的核心-边缘发展空间格局;区域经济总体差异呈现平稳下降的趋势,其中省内差异和河南省内部差异是造成总体差异变动的主要原因;区域经济发展速度极化现象明显,相对发展速度随经济的发展出现分异。

基于NPP-VIIRS夜间灯光数据的山西省GDP空间化模拟

选取山西省为研究对象,以美国国家极轨合作仪件-可见红外成像辐射计套件(NPP-VIIRS)夜间灯光数据、GDP统计数据等为数据源,构建GDP空间化拟合模型,建立山西省GDP密度图,据此研究山西省经济的空间差异性。通过对NPP-VIIRS夜间灯光数据的空间化处理,提取灯光指数,并将其与GDP进行回归拟合,建立最佳回归模型,得到GDP密度拟合图;利用县级GDP数据进行线性纠正,从而提高GDP的模拟精度。结果表明:(1)NPP-VIIRS夜间灯光数据与GDP的相关性较高,可用于山西省GDP模拟;(2)与GDP分区建模相比,GDP整体建模的精度更高;(3)山西省GDP的空间分布整体呈由城市中心逐渐向周边辐射的特点,构成GDP过渡带。

基于NPP-VIIRS夜间灯光数据的京津冀城市群经济集聚特征分析

分析城市群经济集聚特征对识别城市群空间发展模式以及推进城镇化快速发展具有重要意义.选取2012-2018年京津冀城市群NPP-VIIRS夜间灯光数据,在基础处理并经统计数据对比验证的基础上,采用基尼系数、首位城市集聚度以及市域亮源区地均灯光强度表征京津冀城市群整体及各城市的经济集聚程度,运用核密度估计法、标准差椭圆法和重心转移模型探讨城市经济集聚的时空变化特征.结果表明:①研究期内京津冀城市群整体经济集聚程度呈先上升后下降再上升的趋势,但总体上相对稳定,各城市经济集聚程度随时间变化分层更加明显.②城市群经济集聚呈现稳定的双核—多中心发展空间结构,以京津为中心形成双核中心结构引领、西北—东南走向集中的高经济集聚区,以唐山、石家庄和廊坊为中心形成东北—西南走向的经济集聚第二梯度区,其余城市分布在城市群外围形成低经济集聚发展区.③城市经济集聚在离散趋势上呈收缩状态,空间分布上呈现东北—西南分布格局,时间演变上相对稳定;经济集聚重心无明显的方向特征,在东北和西南方向上小幅摆动.

Assessment of Human Impacts on Vegetation in Built-up Areas in China Based on AVHRR,MODIS and DMSP_OLS Nighttime Light Data,1992–2010

Since the reform and opening-up program started in 1978,the level of urbanization has increased rapidly in China.Rapid urban expansion and restructuring have had significant impacts on the ecological environment especially within built-up areas.In this study,ArcGIS 10,ENVI 4.5,and Visual FoxPro 6.0 were used to analyze the human impacts on vegetation in the built-up areas of 656Chinese cities from 1992 to 2010.Firstly,an existing algorithm was refined to extract the boundaries of the built-up areas based on the Defense Meteorological Satellite Program Operational Linescan System(DMSP_OLS)nighttime light data.This improved algorithm has the advantages of high accuracy and speed.Secondly,a mathematical model(Human impacts(HI))was constructed to measure the impacts of human factors on vegetation during rapid urbanization based on Advanced Very High Resolution Radiometer(AVHRR)Normalized Difference Vegetation Index(NDVI)and Moderate Resolution Imaging Spectroradiometer(MODIS)NDVI.HI values greater than zero indicate relatively beneficial effects while values less than zero indicate proportionally adverse effects.The results were analyzed from four aspects:the size of cities(metropolises,large cities,medium-sized cities,and small cities),large regions(the eastern,central,western,and northeastern China),administrative divisions of China(provinces,autonomous regions,and municipalities)and vegetation zones(humid and semi-humid forest zone,semi-arid steppe zone,and arid desert zone).Finally,we discussed how human factors impacted on vegetation changes in the built-up areas.We found that urban planning policies and developmental stages impacted on vegetation changes in the built-up areas.The negative human impacts followed an inverted′U′shape,first rising and then falling with increase of urban scales.China′s national policies,social and economic development affected vegetation changes in the built-up areas.The findings can provide a scientific basis for municipal planning departments,a decision-making reference for government,and scientific guidance for sustainable development in China.

基于SNPP-VIIRS夜间灯光数据的GDP预测模型对比

由于基于夜光遥感数据的GDP预测多采用单一函数模型,缺乏对比和可行性验证,因此利用校正后的夜光影像对比多种函数模型的预测效果具有实际价值。鉴于此,选择SNPP-VIIRS作为基础数据,经过消除负值和背景噪声等校正处理,对线性函数等6种预测模型进行GDP估算效果对比。结果表明:在省级尺度下,二项式函数拟合优度最高,Gaussian函数预测效果最好;市级尺度下,二项式函数均优于其他函数模型。校正方法合理科学,校正后的影像可用于GDP预测研究。

基于NPP-VIIRS夜间灯光数据和Landsat-8数据的城镇建筑用地提取方法改进——以广州市为例

以广州市为例,将Landsat-8-OLS数据与NPP-VIIRS夜间灯光数据相结合提取城镇建筑用地,在常规的NDBI指数、ULI指数方法的基础上进行改进,以获得更加精确的城镇建筑用地信息。首先,利用NPP-VIIRS夜间灯光数据得到城镇建筑用地的大致分布范围,基于Landsat-8-OLS数据提取NDBI、NDVI指数;然后通过分析夜间灯光建筑用地范围、NDVI指数,NDBI指数三者的特性,分别对三者图像进行二值化、求交运算来提取城镇建筑用地;最后将提取结果与传统的NDBI指数、ULI指数方法的提取结果进行对比,利用Google Earth高空间分辨率影像数据进行精度验证。结果表明,结合NPP-VIIRS数据和Landsat-8-OLS数据的多指数提取相交运算,可有效剔除提取结果中包含的裸土信息,与NDBI等指数方法相比,能得到更高的精度。