无人机遥感在应对地质灾害中的主要应用

无人机遥感技术在近几十多年取得了巨大发展,从理论研究已经进入了实用化阶段,尤其是在预防与应对自然灾害方面有着明显优势。介绍了无人机遥感系统的组成、数据获取和处理过程,对地质灾害监测、应急救援和灾情评估工作中无人机遥感可提供的的应用进行了分析,并介绍了无人机遥感在重庆市武隆县鸡尾山特大型滑坡救援中的应用;并对无人机遥感系...     

基于国产遥感数据的秭归县地质灾害区域危险性评价

以位于三峡库区的秭归县为研究区,选择国产2 m全色高分辨率遥感数据进行地质灾害范围信息提取,选用国产HJ-1A/B CCD数据提取NDVI、土地利用分类、土壤含水量等动态地质灾害致灾因子,结合坡度、坡向、降水量、沟壑密度等静态因子,采用单因素叠加的信息量模型进行了区域地质灾害危险性评价,对利用国产遥感数据进行区域地质灾害信息提取和评价进行了有效探索。     

江苏省海岸线演变及地质灾害遥感分析

江苏省漫长的海岸曾经历了大淤大蚀的沧桑变化。现代海岸仍经常受到侵蚀崩岸、岸滩淤涨以及沿岸泥沙流的袭扰破坏。沿岸港口建设、交通航运及滩涂围垦等也不时受到危害,成为江苏海岸带特有的地质灾害。     

无人机红外传感器改制及数据获取研究

针对云雾复杂气候条件下无人机可见光影像难以获取的问题,文章在分析了地质灾害应急遥感监测伴随的复杂气候条件的基础上,根据目前无人机传感器的构造原理,通过对各种红外滤片的分析和比较,研究基于红外滤片元器件的无人机传感器红外改制技术和红外影像数据增强方法,对航摄传感器进行红外改装。实验结果表明,无人机红外传感器在多云雾条件下能够获取高精度清晰影像,有效解决云雾条件下地质灾害应急数据获取难题,为地质灾害应急决策提供参考。     

国外灾害遥感应用研究现状

灾害遥感(调查和监测)是遥感技术应用及研究的一个重要方面。本文根据大量的资料及笔者在国外参加的工作,对经济发达国家及部分第三世界国家将遥感技术用于灾害监测、预报、预警以及在减少灾害损失方面所进行的主要工作及取得的效益作一概略介绍。     

应用遥感图像进行宝成铁路沿线地质灾害的分析

本文通过对4400多张不同时期航片和5幅卫片(部分用计算机和光学图像处理等技术进行了 信息增强提取)的解译, 结合前人资料和地面调查验证, 对宝成铁路(宝鸡-绵阳段)沿线约 5000km²范围的灾害地质现象和环境地质因素进行了分析, 论述了地质灾害的分布发育规律及与地 质环境的相互关系, 提出了铁路沿线地质灾害的防治对策和建议。     

地质灾害遥感调查与监测中高分二号卫星数据应用研究——以湖南新化示范区为例

高分二号成功发射,标志着我国已可进行1 m空间分辨率的遥感监测。为了使高分二号数据能更快地在地质灾害调查领域得到推广,本文针对地质灾害调查与监测中高分二号遥感数据的处理方法进行研究。试验表明,通过对示范区高分二号影像的处理,可获得满足1∶10 000比例尺的平面精度要求的正射影像图,且结果非常适合地质灾害遥感调查与监测。经过多次试验,采用Band1、Band4×0.25+Band2×0.75、Band3组合方法效果较佳,基于ENVI软件的Gram-Schmidt Spectral Sharpening算法进行融合效果较好。     

基于ALOS遥感影像的地质灾害调查研究

以ALOS遥感影像为基础,经过数据处理,结合相关地质资料,对研究区的地层岩性、断裂构造进行了解译,重点解译了工作区内的滑坡、泥石流、崩塌、不稳定边坡等地质灾害,经过野外验证,取得了很好的结果,对研究区排查地质灾害起到了重要作用。此外,该方法对相关工程中地质灾害的调查也具有一定的指导意义。     

湖南省地质灾害遥感应急体系建设研究

本文在总结多年来湖南地质灾害遥感应急调查现状与需求的基础上,结合湖南地质灾害的特点,提出湖南地质灾害遥感应急体系的构建思路,规定了卫星遥感、航空遥感、地面遥感等不同平台的地质灾害遥感应急启动条件和响应时间,详细阐述了灾前基础地理、遥感影像、资源环境和孕灾背景等基础数据库的建设方法以及灾时遥感数据的实时获取途径与处理要求,明确了遥感应急解译与灾情险情评估内容,为湖南省地质灾害遥感应急体系建设提供了重要参考.     

遥感解译在新疆策勒县地质灾害调查中的应用

新疆策勒县地处低-中山区,地质灾害频发.本文首先以RapidEye和ASTER GDEM数据为数据源,在建立遥感解译标志的基础上,采用计算机自动提取和人工判读相结合,遥感解译结果与野外验证相结合的方法,进行了策勒县遥感地质灾害解译.共解译出各类地质灾害点108处,其中崩塌60处、滑坡27处、泥石流21处.其次,利用Ar...     

遥感技术在图们地区地质灾害危险性评价中的应用

图们市是地质灾害多发地区,这些地质灾害具有密集性、易发性、群发性和重复性的特点,对当地人民生命财产安全构成威胁,更制约着当地社会经济可持续发展。本次研究以2005年SPOT-5影像、2015年GF-2影像为数据源,选取地貌、岩土体类型、地质灾害点密度、地质构造、森林覆盖率、降雨量和人类工程活动7个评价因子,采用网格分析法获取各因子的量化分级图。通过Arc GIS的栅格计算功能,将各因子加权叠加获取危险性指数分布图,再通过surfer重分类生成研究区地质灾害危险性区划图。研究表明,图们市地质灾害危险性程度以轻度和微度为主,重度和中度区主要集中在重要交通干线和居民集中生活区,应加强危险性重度和中度区的地质灾害监测和排查工作。     

九寨沟核心景区多源遥感数据地质灾害解译初探

点云密度是激光雷达（light detection and ranging,LiDAR）技术的重要参数,对森林遥感反演指数的提取有重要影响。以1 600 m×1 450 m大小的无人机（unmanned aerial vehicle,UAV）LiDAR数据为实验数据,采用分级随机抽稀法对实验数据进行抽稀,获取不同密度的点云数据集,利用不同密度数据集提取郁闭度、间隙率、叶面积指数、点云高度和密度分位数等森林遥感反演指数,并与原始数据提取的森林遥感反演指数进行差值比较。（1）当点云密度较小时,提取的郁闭度略微偏低,而间隙率略微增加,点云密度对郁闭度、间隙率的影响极小。（2）当点云密度较大时,对叶面积指数的影响不大,但当点云密度较小时,对叶面积指数的影响较大,个别区域可能出现叶面积指数突变。（3）当点云密度较大时,点云密度对高度、密度分位数的影响不明显,但当点云密度降至3.6点/m²时,可能会出现个别区域密度、高度密度分位数突变的情况。点云密度对森林遥感反演指数有重要影响,合适的点云密度有利于更准确地描述森林结构形态,过小的点云密度影响森林遥感反演指数的提取。

     

重大地质灾害隐患早期识别中综合遥感应用的思考与建议

青藏高原交通廊道作为中国意义非凡的重大工程,也是世界铁路史上地形最复杂、地质安全问题最严峻的铁路。合成孔径雷达干涉测量技术（interferometric synthetic aperture radar,InSAR）凭借其广覆盖、高精度、高效率等特点已成为地质灾害隐患早期识别与监测的重要手段,但是其在青藏高原地形起伏剧烈的高山峡谷区域会面临较为严重的几何畸变问题。以覆盖面最广、获取时间最密的全球完全开放获取的哨兵1号卫星（Sentinel-1）数据为例,将青藏高原交通廊道沿线作为研究区域开展卫星雷达影像几何畸变与适宜性分析。综合考虑合成孔径雷达（synthetic aperture radar,SAR）成像几何、Sentinel-1影像斜距向入射角差异及被动几何畸变3个方面,提出了一套基于局部入射角的Sentinel-1影像几何畸变区域精准定量识别方法,并对全线几何畸变进行了定量计算与升降轨适宜性分析。结果显示,青藏高原交通廊道所在极陡高山峡谷区单轨道几何畸变区域达到31%~35%,升降轨联合观测可以有效将几何畸变区域减小至1.5%,同时约有35%区域适宜升降轨同时观测进行联合分析。

     

Remote Sensing Data Acquisition,Platforms and Sensor Requirements

Although data available from various earth observation systems have been routinely used in many resource applications, however there have been gaps, and data needs of applications at different levels of details have not been met. There is a growing demand for availability of data at higher repetivity, at higher spatial resolution, in more and narrower spectral bands etc. Some of the thrust areas of applications particularly in the Indian context are;

1. Management of natural resources to ensure sustainable increase in agricultural production,
2. Study the state of the environment, its monitoring and assessment of the impact of. various development actions on the environment,
3. Updating and generation of large scale topographical maps.
4. Exploration/exploitation of marine and mineral resources and
5. Operational meteorology and studying various land and oceanic processes to understand/predict global climate changes.

Each of these thrust area of application has many components, related to basic resource areas such as agriculture, forestry, water resources, minerals, marine resources etc. and the field of cartography. Observational requirements for major applications have been summarized as under. Monitoring vegetation health from space remains the most important observational parameter with applications, in agriculture, forestry, environment, hydrology etc. Vegetation extent, quantity and temporal changes are the three main requirements which are not fully realized with RS data available. Vegetation productivity, forest biomass, canopy moisture status, canopy biogeochemistry are some examples. Crop production forecasting is an important application area. Remotely sensed data has been used for identification of crops and their acreage estimation. Fragmented holdings, large spread in crop calendars and different management practices continue to pose a challenge lo remote sensing. Remotely sensed data at much higher spatial resolution than hitherto available as well as at greater repetivity are required to meet this need. Non-availability of cloud-free data in the kharif season is one of the serious problems in operational use of remote sensing for crop inventory. Synthetic aperture radar data al X & Ku bands is necessary to meet this demand. Nutrient stress/disease detection requires observations in narrow spectral bands. In case of forestry applications, multispectral data at high spatial resolution of the order of 5 to 10 metres is required to make working plans at forest compartment level. Observations from space for deriving tree height are required for volume estimation. Observations in the middle infrared region would greatly enhance capability of satellite remote sensing in forest fire detection. Temporal, spatial and spectral observational requirements in various applications on vegetation viewing are diverse, as they address processes at different spatial and time scales. Hence, it would be worthwhile to address this issue in three broad categories. a) Full coverage, moderate spatial resolution with high repetivity (drought, large scale deforestation, forest phenology....). b) Full coverage, moderate to high spatial resolution and high repetivity (crop forecasting, vegetation productivity). c) Selected viewing at high spatial resolution, moderate to high repetivity and with new dimensions to imaging (narrow spectral bands, different viewing angles). A host of agrometeorological parameters are needed to be measured from space for their effective use in development of yield models. Estimation of root-zone soil moisture is an important area requiring radar measurements from space. Surface meteorological observations from space at the desired spatial and temporal distributions has not developed because of heavy demands placed on the sensor as well as analytical operational models. Agrometeorology not only provides quantitative inputs to other applications such as crop forecasting, hydrological models but also could be used for farmer advisory services by local bodies. Mineral exploration requires information on geological structures, geomorphology and lithology. Surface manifestation over localized regions requires large scale mapping while the lithology can be deciphered from specific narrow bands in visible. NIR, MIR and TIR regions. Sensors identified for mapping/cartography in conjunction with imaging spectrometer would seem to cover requirements of this application. Narrow spectral bands in the short regions which provide diagnostics of relevant geological phenomenon are necessary for mineral exploration. Thermal inertia measurements help in better discrimination of different rock units. Measurements from synthetic aperture data which would provide information on geological structures and geomorphology are necessary for mineral exploration. The applications related to marine environment fall in three major areas: (i) Ocean colour and productivity, biological resources; (ii) Land-ocean interface, this includes coastal landforms, bathymetry, littoral transport processes, etc. and; (iii) Physical oceanography, sea surface temperature, winds, wave spectra, energy and mass exchange between atmosphere and ocean. Measurement of chlorophyll concentration accurately on daily basis, sea surface temperature with an accuracy of 0.5 °K. and information on current patterns arc required for developing better fishery forecast models. Improved spatial resolution data are desirable for studying sediment and other coastal processes. Cartography is another important application area. The major problems encountered in relation to topographic map updation are location and geometric accuracy and information content. Two most important requirements for such an application are high spatial resolution data of 1 to 2 metre and stereo capability to provide vertical resolution of 1 metre. This requirement places stringent demands on the sensor specifications, geometric processing, platform stability and automated digital cartography. The requirements for the future earth observation systems based on different application needs can be summarized as follows:

1. Moderate spatial resolution (l50-300m), high repetivity (2 Days), minimum set of spectral bands (VIS, NIR, MIR. TIR) full coverage.
2. Moderate to high spatial resolution (20-40m), high repetivity (4-6 Days), spectral bands (VIS, MR, MIR, TIR) full coverage.
3. High spatial resolution (5-10m) muitispectral data with provision for selecting specific narrow bands (VIS, N1R. MIR), viewing from different angles.
4. Synthetic aperture radar operating in at least two frequencies (C, X, Ku), two incidence angles/polarizations, moderate to high spatial resolution (20-40m), high repetivity (4-6 Days).
5. Very high spatial resolution (1-2m) data in panchromatic band to provide terrain details at cadastral level (1:10,000).
6. Stereo capability (1-2m height resolution) to help planning/execution of development plans.
7. Moderate resolution sensor operating in VIS, NIR, MIR on a geostationary platform for observations at different sun angles necessary for the development of canopy reflectance inversion models.
8. Diurnal (at least two i.e. pre-dawn and noon) temperature measurements of the earth surface.
9. Ocean colour monitor with daily coverage.
10. Multi-frequency microwave radiometer, scatterometer. altimeter, atmospheric sounder, etc.

     

遥感技术在三亚海岸带地质环境调查中的应用

为了研究三亚海岸带地质环境状况,本文通过遥感技术,采用1987年的TM、2000年的TM、2010年的Rapideye和2015年的QB2四个时相的卫星遥感数据,对比研究过去28年来三亚海岸带地质环境变化情况,利用平衡岬湾模型软件Mepbay对工程用海海岸进行静态平衡岸线模拟,在研究中发现三亚海岸带地质环境有较大变化,地质灾害点较多,岸线侵蚀、淤积问题突出,岸线变迁幅度较大,造成三亚海岸带地质环境变化的主要原因是人们过度开发海岸带资源,用海工程规划设计不合理,缺乏环境保护意识,其次是自然因素。     

中亚天山地区横向构造的航天遥感分析

运用中小比例尺航天图象。对中亚天山地区的横向构造进行研究，探讨深部构造信息的地表遥感特征及深部不均一性与地表不均一性的关系，并分析了天山地区大－超大型金、铜矿床的遥感地质背景。     

遥感地质信息提取集成与矿物遥感地质分析模型

结合遥感技术的现状、发展趋势和资源环境应用需求，从遥感地质应用出发，提出多源遥感数据分层识别集成、基于岩矿识别技术体系集成和基于岩矿识别谱系集成的基本思路与框架，强调矿物或矿物集合体在遥感地质分类与识别中的重要性，初步建立矿物遥感地质分析模型，分析集成与模型在蚀变与矿化信息提取、地质环境评价等中的可应用性。     

重大地质灾害隐患早期识别中综合遥感应用的思考与建议

2017年以来,以四川茂县新磨村高位垮塌、金沙江白格滑坡为代表的重大地质灾害多次发生,呈现出隐蔽性强、突发性高、破坏力大等特点,难以早期发现。值得注意的是,尽管地质灾害防治工作已在全国大范围开展并受到国家层面的高度重视与支持,但绝大部分灾害仍发生于全国现有地质灾害隐患点分布以外的区域。对这些人迹罕至、人不能至且观测条件苛刻的地区的重大隐患进行有效早期识别,是当前地质灾害防治工作尚需解决的难题和重要任务。在总结合成孔径雷达干涉测量（interferometric synthetic aperture radar,InSAR）技术特点与其应用局限的基础上,从光学遥感、InSAR、激光雷达等综合遥感测量的角度提出了以“形态、形变、形势”（三形）为观测内容、以定性识别灾害隐患位置、定量监测灾害体变形幅度、依靠综合遥感动态监测数据提升隐患识别能力的技术思路。对未来工作提出了若干建议与思考,以期服务于重大地质灾害隐患综合判断与早期识别工作。     

遥感技术在地质构造及找矿中的应用

利用Landsat-8卫星的ETM+遥感影像为数据源,结合当地区域地质调查基础资料,对中国西北某高原地区内线性断裂、环形构造、侵入岩体、赋矿地层等地质构造的成矿特征开展遥感解译,综合运用数理统计原理与地统计学分析方法,分析遥感解译的地质构造信息与矿产勘查的相关性,总结了区内解译构造与成矿关系条件。     

鄱阳湖地区高光谱遥感数据的定标研究

针对鄱阳湖地区的成像光谱仪遥感数据,利用3种不同的定标模型进行了定标反演的研究。结果表明,在所有的3种定标方法中,经验线性定标模型的反演结果最稳定,基本符合实际。