复杂卫星图像中的小目标船舶识别

船舶作为海上的重要目标,实现对船舶自动识别有重要的意义。针对卫星图像中云雾、海岸背景等复杂海情对船舶识别带来的干扰,以及小目标船舶高漏检率问题,本文提出一种多尺度深度学习模型训练策略,在此基础上构建了一种船舶识别的深度学习网络,该网络可分为多尺度训练、特征提取、生成目标建议区域、船舶分类这4个部分。首先,采用多尺度的训练策略,将多尺度的船舶样本送入网络中进行训练,这样在训练样本中加入了大量小目标船舶的样本,使网络充分提取到小目标船舶的特征;其次,通过卷积神经网络对目标船舶进行特征自适应提取;然后,目标区域建议网络可依据卷积神经网络提取到的特征,在图像中找到感兴趣目标区域,即框定船舶的位置;最后,通过多个全连接层的组合,将高维特征映射到一个4元组中,再运用分类函数输出每一类船舶的概率值,概率值最大的则为该船舶的类别。同时为解决云雾遮挡和海岸背景的干扰,采用了一种负样本增强学习的方法,在样本数据集中加入了大量只含有云雾和海岸背景的图片,进行负样本扩充,增强网络模型对云雾及海岸背景的特征学习能力,以此解决复杂海情的影响。实验结果表明,所提方法有效解决了复杂海情条件下的船舶识别难,以及小目标船舶识别难的问题,实现了复杂海情条件下的船舶识别。同时,与现有成熟的深度学习目标识别算法相比,本文算法的精确度和召回率分别提升了6.98%和18.17%,所训练的模型具有良好的泛化能力和鲁棒性。     

新冠病毒疫情期间复工复产卫星遥感监测

本文利用美国新一代极轨卫星NPP搭载的VIIRS载荷数据反演获得的热点数据,以及美国NASA的Aura卫星搭载的OMI载荷的二氧化氮对流层柱浓度资料,对中国地区2020-01—02及2019年同期,以及抗击疫情期间和后期复工复产的情况进行监测。监测结果表明：（1）2020-02较1月份及2019年同期相比,工业企业的耗能水平却出现显著降低,NO₂浓度呈现显著下降趋势。（2）全国工业热源点数和分布范围及密度自2020-02-03正式复工后稳步提升。湖北省除武汉、黄石和鄂州地区仍可监测到少数热点外,其他地区2月份均未再监测到任何热点,也证明了湖北基本没有复工迹象。（3）疫情得到基本控制之后,京津冀地区工业热源点有显著增加,但新增区域一般仍处于低能耗水平。从全国大气NO₂柱浓度监测来看,复工开始前后与2019年同期相比,污染物浓度均呈现明显下降趋势,说明复工强度有限。（4）到3月初工业热源点及其能耗水平出现显著升高,中国大部分地区的热源企业虽已逐步复工生产,说明工业产能较2月有显著提升,但生产规模或产能并未完全恢复。     

Geospatial technologies for Chang’e-3 and Chang’e-4 lunar rover missions

ABSTRACT

This paper presents a brief overview of the geospatial technologies developed and applied in Chang’e-3 and Chang’e-4 lunar rover missions. Photogrammetric mapping techniques were used to produce topographic products of the landing site with meter level resolution using orbital images before landing, and to produce centimeter-resolution topographic products in near real-time after landing. Visual positioning techniques were used to determine the locations of the two landers using descent images and orbital basemaps immediately after landing. During surface operations, visual-positioning-based rover localization was performed routinely at each waypoint using Navcam images. The topographic analysis and rover localization results directly supported waypoint-to-waypoint path planning, science target selection and scientific investigations. A GIS-based digital cartography system was also developed to support rover teleoperation.     

An IEEE value loop of human-technology collaboration in geospatial information science

ABSTRACT

Geosensing and social sensing as two digitalization mainstreams in big data era are increasingly converging toward an integrated system for the creation of semantically enriched digital Earth. Along with the rapid developments of AI technologies, this convergence has inevitably brought about a number of transformations. On the one hand, value-adding chains from raw data to products and services are becoming value-adding loops composed of four successive stages – Informing, Enabling, Engaging and Empowering (IEEE). Each stage is a dynamic loop for itself. On the other hand, the “human versus technology” relationship is upgraded toward a game-changing “human and technology” collaboration. The information loop is essentially shaped by the omnipresent reciprocity between humans and technologies as equal partners, co-learners and co-creators of new values.

The paper gives an analytical review on the mutually changing roles and responsibilities of humans and technologies in the individual stages of the IEEE loop, with the aim to promote a holistic understanding of the state of the art of geospatial information science. Meanwhile, the author elicits a number of challenges facing the interwoven human-technology collaboration. The transformation to a growth mind-set may take time to realize and consolidate. Research works on large-scale semantic data integration are just in the beginning. User experiences of geovisual analytic approaches are far from being systematically studied. Finally, the ethical concerns for the handling of semantically enriched digital Earth cover not only the sensitive issues related to privacy violation, copyright infringement, abuse, etc. but also the questions of how to make technologies as controllable and understandable as possible for humans and how to keep the technological ethos within its constructive sphere of societal influence.     

Atmospheric pollution assessment near potential source of natural aerosols in the South Gobi Desert region,China

ABSTRACT

Rapid economic growth, a high degree of urbanization and the proximity of a large number of desert and semidesert landscapes can have a significant impact on the atmosphere of adjacent territories, leading to high levels of atmospheric pollution. Therefore, identifying possible sources of atmospheric pollution is one of the main tasks. In this study, we carried out an analysis of spatial and temporal characteristics of five main atmospheric pollutants (PM_2.5, PM₁₀, SO₂, NO₂, and CO) near potential source of natural aerosols, affecting seven cities (Wuhai, Alashan, Wuzhong, Zhongwei, Wuwei, Jinchang, Zhangye), located in immediate proximity to the South Gobi deserts. The results, obtained for the period from 1 January 2016 to 31 December 2018, demonstrate total concentrations of PM_2.5 and PM₁₀ are 38.2 ± 19.5 and 101 ± 80.7 μg/m³ exceeding the same established by the Chinese National Ambient Air Quality Standard (CNAAQS), being 35 and 70 μg/m³, respectively. Based on the data from Moderate Resolution Imaging Spectroradiometer (MODIS) for the whole period, Clean Сontinental (71.49%) and Mixed (22.29%) types of aerosols prevail in the region. In the spring and winter seasons maximum concentrations of pollutants and high values of Aerosol Optical Depth (AOD) in the region atmosphere are observed. PM_2.5 and PM₁₀ ratio shows the presence of coarse aerosols in the total content with value 0.43. The highest concentrations of pollutants were in the period of dust storms activity, when PM_2.5 and PM₁₀ content exceeded 200 and 1000 µg/m³, and AOD value exceeded 1. UV Aerosol Index (UVAI), Aerosol Absorbing Optical Depth (AAOD), and Single Scattering Albedo (SSA), obtained from Ozone Monitoring Instrument (OMI), demonstrate the high content of dust aerosols in the period of sandstorms. Analysis of backward trajectories shows that dust air masses moved from North to Northwest China, affecting large deserts such as Taklamakan, Gurbantunggut, Badain Jaran, Tengger, and Ulan Buh deserts.     

Combining European Earth Observation products with Dynamic Global Vegetation Models for estimating Essential Biodiversity Variables

ABSTRACT

Global, fast and accessible monitoring of biodiversity is one of the main pillars of the efforts undertaken in order to revert it loss. The Group on Earth Observations Biodiversity Observation Network (GEO-BON) provided an expert-based definition of the biological properties that should be monitored, the Essential Biodiversity Variables (EBVs). Initiatives to provide indicators for EBVs rely on global, freely available remote sensing (RS) products in combination with empirical models and field data, and are invaluable for decision making. In this study, we provide alternatives for the expansion and improvement of the EBV indicators, by suggesting current and future data from the European Space Agencýs COPERNICUS and explore the potential of RS-integrated Dynamic Global Vegetation Models (DGVMs) for the estimation of EBVs. Our review found that mainly due to the inclusion of the Sentinel constellation, Copernicus products have similar or superior potential for EBV indicator estimation in relation to their NASA counterparts. DGVMs simulate the ecosystem level EBVs (ecosystem function and structure), and when integrated with remote sensing data have great potential to not only offer improved estimation of current states but to provide projection of ecosystem impacts. We suggest that focus on producing EBV relevant outputs should be a priority within the research community, to support biodiversity preservation efforts.     

一种遥感信息服务分类本体构建方法

在分析遥感信息服务的领域特征及其概念的事件名词特点的基础上,结合框架语义学理论,提出了基于事件框架提取遥感信息服务概念的语义特征并在此基础上利用形式概念分析构建遥感信息服务分类本体的方法。实验结果表明了此方法的有效性。     

数字地球中影像数据的Zipf-like访问分布及应用分析

在数字地球GlobeSIGht系统中,统计分析不同采集时间尺度的4个系统日志,得出了影像数据瓦片请求均符合幂律分布中的Zipf-like分布的结论。该结论有助于数字地球系统在诸如影像数据存储分布和缓存等方面的性能优化设计和分析。     

基于CityGML的遥感信息共享研究

在参考现有地物分类标准和CityGML要素类型基础上,探讨遥感专题模型的设计方法,分析基于CityGML的几何拓扑、语义属性信息表达机制,并以一小区域数据验证了方法的可行性。