基于Google Earth Engine的湖泊水位与水体面积关系研究

对于传统的湖泊水位-面积关系模型研究, 长时间序列遥感影像水体提取存在数据下载和处理繁琐、大量使用专业软件和人工干预多等问题, 研究过程耗时长、效率低. 基于Google Earth Engine, 对2002年至2016年长时间序列Landsat影像数据进行预处理, 利用NDWI提取东洞庭湖水体范围并计算水域面积, 结合城陵矶水文站日水位数据, 设计水位加权平均法、阈值法和单点水位法比较并确定影像对应的最优水位, 建立水位-水体面积关系曲线. 研究结果表明: (1)东洞庭湖水域范围随季节变化大, 面积在7～9月最大, 在1～3月最小; (2)利用阈值法计算最优水位的精度(R2=0.9628)优于加权平均水位法(R2=0.8322)和单点水位法(R2=0.9457); (3)五次多项式模型建立水位-面积关系曲线拟合精度最高(R2=0.9628).     

光学和SAR遥感数据在赵村水库水体监测中的应用

为了解决传统水库监测管理手段效率低下的问题，采用光学影像和SAR影像数据对水库水体进行提取，实现对水库水域面积的动态监测。对比分析了多种水体提取方法在赵村水库的适用性，并结合现场调研的实际情况，分析了降雨量、水域面积和水位之间的关系。此外，利用数字高程模型对水库淹没区进行了模拟计算，得出了不同坝高设计下的淹没区域面积，并结合遥感影像进行了可视化展示。研究结果表明，光学影像和SAR影像数据相结合的方法可以准确提取水库水体信息，结果与实地调研数据相符，可为后续的水库监测预警工作提供有效的数据支持。     

基于机载LiDAR 技术的典型水库水量变化估算

针对传统外业测量水库库容估算方法易受地形、气象条件影响,存在危险性高、效率低等问题,以仑山水库为研究对象,采用机载 LiDAR 技术估算水库面积及水量变化。通过实地数据采集,利用 KD- 树算法剔除点云中的粗差点,并采用渐进加密不规则三角网(PTIND)滤波分离出地面点,完成 LiDAR 点云数据的预处理;根据点云数据建立精细化数字高程模型(DEM),提取不同水位的水库水面面积;最终利用积分和传统棱台体积估算方法对比分析不同水位水量差值变化。结果表明：与高程实测值相比,DEM 高程反演值满足高程精度要求,基于机载 LiDAR 的水量估算受库底坡度变化的影响,估算值低于传统估算方法的估算值,估算结果更为精确,可为监测水库水量变化提供科学依据。     

2012～2014年江苏省水域面积变化遥感监测研究

对水域的动态监测,有利于及时掌握水域变化信息,加强水资源的管理和保护。高分辨率遥感对地识别能力强,监测成果真实、客观、准确、全面,是水域动态监测的一种先进的技术手段。本研究以江苏省水域面积为监测对象,利用2012年和2014年两期0.3米空间分辨率航片,开展2012～2014年江苏全省水域面积变化遥感监测研究。经遥感监测,外业调查,数据统计,结果表明：(1)2012～2014年江苏省水域面积总计增加5.457km2,其中南通、连云港、扬州和泰州共4市水域面积减少,而南京、无锡、徐州、常州、苏州、淮安、盐城、镇江和宿迁共9市水域面积增加;(2) 2012-2014年江苏水域面积增加的类型为湖泊和塘坝,减少的类型为河流和水库;(3)江苏省水域面积变化主要原因包括河湖监管力度加大、湖泊退圩还湖、长江码头建设、水库水域岸线开发占用等。通过本研究,利用高空间分辨率遥感影像开展水域面积变化监测,可快速、准确获取大范围区域水域变化信息,为实现水域的动态监管、水行政执法、水资源的管理和保护提供先进的技术手段和基础数据支撑。     

多源遥感技术在土地利用的应用综述

该文简述了多源遥感的概念,多源遥感技术的意义和作用以及多源遥感应用的有关技术(主要包括遥感影像的融合技术)的现状的进展,并着重阐述了多源遥感在土地利用方面的一些进展;同时也提出了些许问题和一些期望。     

基于行道树的高分辨率遥感影像道路提取研究

在城市遥感中,道路自动(半自动)提取一直是研究的重点.本文在现有道路自动(半自动)提取技术基础上,结合高分辨率遥感影像中行道树的分布特点,从地学知识出发,利用归一化植被指数、阀值分割、数学形态学算子及地理信息系统的分析功能等实现基于行道树的道路自动(半自动)提取和道路面积快速、自动计算.结论表明基于行道树的道路提取和面积计算方法具有一定的实用性.     

面向对象多特征融合的水域岸线目标变化检测

为认真落实河长制"清四乱"等专项行动,量化水域岸线监管测评工作,以岸线码头为目标,研究一种基于面向对象思想多特征融合的水域岸线目标变化检测方法。针对多时相高分辨率遥感影像,利用面向对象多尺度分割原理将具有空间连续性的同类区域划分为目标对象,提取目标对象的光谱、纹理及几何结构组成特征矩阵,并利用高斯径向基核函数支持向量机(RBF-SVM)进行分类;计算变化矢量差值,并与人工判读数据对比分析得到目标变化检测结果。实验结果表明,该研究应用于水域岸线上目标的变化检测中效果明显,RBFSVM分类误差影响最终目标变化检测的正确率,可为实现河湖水域岸线长效管护提供技术支撑。     

基于多源数据的白洋淀水域变化分析

以白洋淀水域为研究区域,针对白洋淀水域上游2个河道水文站的年平均流量、水位及水域遥感监测面积等数据,进行Mann-Kendall趋势、突变的检验,以及水位-面积拟合,并对结果进行分析,以期从多源数据中挖掘白洋淀水域的变化特征。分析结果表明:紫荆关站与阜平站的年平均流量呈显著下降趋势,且这种下降趋势在1990年后更加显著;白洋淀的水面面积经过急速下降—持续较低—缓慢上升的发展阶段,整体呈不显著下降状态,并且按先中部东部,再南部北部,最后西部的空间顺序增长;由于白洋淀上游河流流量不足,引水济淀仍然是保持白洋淀正常生态环境的有效途径;白洋淀春季、秋季、冬季水位-面积正相关关系明显,夏季较差,春、秋、冬三季水位-面积整体正相关且拟合关系较好。     

基于图像块SIFT特征学习的多源遥感图像匹配方法北大核心CSCD

对于多源遥感图像因成像原理、时相差异以及分辨率等因素导致的匹配困难问题,提出了一种基于图像块SIFT特征学习的多源遥感图像匹配方法。首先,通过尺度不变特征变化(scale-invariant feature transform,SIFT)提取图像特征点并截取对应的图像块对;其次,利用多源遥感图像匹配网络(matching nerul network,MNN)学习图像块特征并输出匹配点对,结合快速样本一致性(fast sample consensus,FSC)方法优化匹配结果;最后,计算图像变换矩阵实现多源遥感图像配准。为验证本文方法的有效性,制作了8000对多源遥感图像数据集对MNN进行网络训练,并与FSC-SIFT(fast sample consensus-scale-invariant feature transform)、PSO-SIFT(position scale orientation-scale-invariant feature transform)以及跨模态图像匹配网络Contextdesc进行对比实验,结果表明本文方法在正确匹配点数量、匹配精度等方面具有一定优越性。     

鄱阳湖面积的卫星遥感估计及其与水位关系分析

鄱阳湖水位变幅巨大,近年来旱涝灾害严重,对其进行实时监测具有重要意义。MODIS遥感数据时间分辨率很高,在水体动态监测中具有明显优势。水体提取方法多种多样,在比较不同水体提取方法的基础上,最终选用改进的归一化水体指数(MNDWI)的方法进行鄱阳湖水体提取;利用2000~2010年40期MODIS数据分析了鄱阳湖这10 a的水面面积年际演变特征及季节性变化特征,同时结合同时期的水位数据,建立了鄱阳湖面积-水位关系的模型。与地形图量测结果的对比表明,遥感解译的鄱阳湖水体面积-水位关系与根据地形图分析建立的面积-水位关系具有较好的一致性。     

西南地区不同山地环境梯度叶面积指数遥感反演

叶面积指数(LAI)遥感估算是植被定量遥感研究的热点之一,监测植被LAI时空变化对于研究陆地生态系统碳循环及全球变化等具有非常重要的意义。在我国西南山区设置10个50km×50km的观测样区作为研究区,其中包括5个森林生态系统样区、3个农田生态系统样区和2个草地生态系统样区。分别获取不同优势植被类型LAI地面实测数据,结合同期获取的遥感数据,考虑地形因素影响,基于偏最小二乘原理分别构建各样区LAI遥感估算模型,并采用交叉验证的方式对模型精度进行评价。结果表明:考虑了海拔、坡度和坡向等地形因子的森林LAI遥感反演模型与未考虑地形变量的模型相比,其验证精度有所提高,R2由0.30~0.75提高至0.50~0.80,RMSE由0.52~0.93m2/m2降低至0.48~0.89m2/m2;所有样区优势植被类型LAI反演模型验证R2在0.40~0.80之间,RMSE在0.22~0.89m2/m2之间。发展的LAI遥感估算方法有助于认知山地植被LAI反演的地形效应问题,可为进一步的山地植被长势监测提供科学依据。     

突发水污染事件中遥感瓦片大数据存储系统

突发水污染事件频发,利用GIS平台、遥感技术及水质模型对突发水污染事件进行模拟研究逐渐成为一种发展趋势.针对目前海量、异构、动态遥感数据难以快速处置和高效存储的问题,基于HBase设计和实现了突发水污染事件的遥感瓦片大数据存储系统.该系统以数字地球平台为支撑,结合遥感瓦片金字塔模型和MapReduce框架,考虑实时、动态观察事件发展态势引起的瓦片数据频繁加载特点,对遥感瓦片数据的线性四叉树编码索引进行了改进,将其同水污染数据一并存储到HBase数据库中,并通过增加缓存区来提升遥感瓦片数据的访问效率.将此系统应用于渭河陕西段突发水污染事件中,实验结果表明:改进后的索引可在200个毫秒时间内提取到水污染区域的遥感数据,引入的缓存机制使得水污染区域遥感数据的加载时间节省了近2/3.因此,该存储系统访问效率高,能够满足突发水污染事件快速应急模拟的需求.     

红树林雷达遥感研究进展

红树林是海岸带生态系统中重要的植物群落,具有较高的社会、生态和经济价值。遥感技术的发展为红树林监测提供了一种高效便利的手段。雷达遥感由于具有穿透性好、不受云雨影响的特点,在红树林分布地区具有得天独厚的优势。对近几十年来雷达遥感在红树林监测方面的研究进行了回顾,着重分析了红树林散射机制、红树林分类与识别和红树林生物物理参数反演这3个方面的研究进展,对各类方法进行了总结和对比,最后针对存在的问题提出了未来可以改进的方向。     

Change Detection of Han Hangu Pass World Heritage Site based on PALSAR-1 Remote Sensing Images

Due to its data acquisition capacity in all\|weather and all\|time conditions,Synthetic Aperture Radar(SAR) technology has been a valuable tool in change detection,implying the potential in various applications.In this paper,we jointly applied SAR coherent information and intensity information for the change detection (primarily caused by the demolition behavior in 2010~2011)in the study site of Han Hangu Pass (a World Heritage site)using seven PALSAR\|1 remote sensing images acquired in the period from 2007 to 2011.Results show that the two change detection methods,either based on coherent information or intensity information,yielded reliable results,with the detection probability of 0.868 and 0.697,respectively,in conjunction with the false\|alarm probability of 0.385 and 0.197.The comparison related to two methodsindicated that the coherence of PALSAR\|1 was more sensitive to the surface change than the other owing to its higher signal\|to\|noise ratio.Limited by the negative impacts of speckle and low\|median resolution,the application of intensity\|based change detection using PALSAR\|1 still remains problematic.     

Factors affecting remotely sensed snow water equivalent uncertainty

State-of-the-art passive microwave remote sensing-based snow water equivalent (SWE) algorithms correct for factors believed to most significantly affect retrieved SWE bias and uncertainty. For example, a recently developed semi-empirical SWE retrieval algorithm accounts for systematic and random error caused by forest cover and snow morphology (crystal size — a function of location and time of year). However, we have found that climate and land surface complexities lead to significant systematic and random error uncertainties in remotely sensed SWE retrievals that are not included in current SWE estimation algorithms. Joint analysis of independent meteorological records, ground SWE measurements, remotely sensed SWE estimates, and land surface characteristics have provided a unique look at the error structure of these recently developed satellite SWE products. We considered satellite-derived SWE errors associated with the snow pack mass itself, the distance to significant open water bodies, liquid water in the snow pack and/or morphology change due to melt and refreeze, forest cover, snow class, and topographic factors such as large scale root mean square roughness and dominant aspect. Analysis of the nine-year Scanning Multichannel Microwave Radiometer (SMMR) SWE data set was undertaken for Canada where many in-situ measurements are available. It was found that for SMMR pixels with 5 or more ground stations available, the remote sensing product was generally unbiased with a seasonal maximum 20 mm average root mean square error for SWE values less than 100 mm. For snow packs above 100 mm, the SWE estimate bias was linearly related to the snow pack mass and the root mean square error increased to around 150 mm. Both the distance to open water and average monthly mean air temperature were found to significantly influence the retrieved SWE product uncertainty. Apart from maritime snow class, which had the greatest snow class affect on root mean square error and bias, all other factors showed little relation to observed uncertainties. Eliminating the drop-in-the-bucket averaged gridded remote sensing SWE data within 200 km of open water bodies, for monthly mean temperatures greater than − 2 °C, and for snow packs greater than 100 mm, has resulted in a remotely sensed SWE product that is useful for practical applications.     

A Review on Estimating Fractional Cover of Non-photosynthetic Vegetation by Using Remote Sensing

The quantitative estimation of fractional cover of photosynthetic vegetation(f PV),non-photosynthetic vegetation(f NPV),and bare soil(f BS) is critical for grassland ecosystem carbon storage,vegetation productivity,soil erosion and wildfire monitoring.The ecological importance of NPV has driven considerable research on quantitatively estimating NPV in diverse ecosystems including croplands,forests,grasslands savannah,and shrublands using remote sensing.This paper reviews the research progress in estimating f NPV using hyperspectral and multisspcetral remote sensing data,and hightlights discusses the theoretical bases of PV,NPV and BS spectral characteristics.based on the existing methods for estimating f NPV,this article groupd into two categories:empirical relationship between spectral index and NPV cover,and Spectral mixture analysis.Meanwhile,also discuss applications.of hyperspectral and multisspcetral remote sensing data.Finally,the existential problems and research trends for NPV estimation are analyzed.     

Remote sensing support for national forest inventories

National forest inventory programs are tasked to produce timely and accurate estimates for a wide range of forest resource variables for a variety of users and applications. Time, cost, and precision constraints cause these programs to seek technological innovations that contribute to measurement and estimation efficiencies and that facilitate the production and distribution of an increasing array of inventory data, estimates, and derived products. Many of the recent innovations have involved remotely sensed data and related statistical estimation techniques. Current applications of remote sensing in support of national forest inventories are reviewed for three areas: (1) observation or measurement, meaning using remotely sensed data in lieu of field observations or measurements; (2) estimation, meaning calculation of traditional inventory areal estimates such as forest area or volume per unit area; and (3) mapping. Future applications focus on two areas: augmenting field measurements with remotely sensed data obtained from lidar sensors and Internet accessible map-based estimation.     

积雪遥感动态研究的现状及展望

简要讨论了积雪遥感研究的现状,主要包括常用传感器的物理参数及其可行性和局限性,云和雪的区分技术,雪盖面积和积雪深度的提取,雪水当量换算以及积雪遥感在融雪径流模拟、雪灾监测与评价、积雪对气候变化的影响研究等方面的应用。并对积雪遥感研究的发展趋势做了简要的分析与展望。     

Satellite‐based assessment of the dynamics of new lakes in southern Egypt

Lakes in arid landscapes are indicators of environmental change and important sources of water for human use. In regions without in situ hydrologic measurements, remote sensing may provide the only means to monitor long‐term changes in water storage. We used a synergistic combination of multiple satellite remote‐sensing methods to provide the first comprehensive assessment of the dynamics of a newly formed chain of large lakes in the hyper‐arid Toshka Depression of southern Egypt. A total of 145 MODIS and AVHRR satellite images were used to monitor changes in lake surface area, which increased to a maximum of 1740 km² before declining to 900 km². Two methods were tested for satellite‐based measurement of lake levels and volumes, one based on analysis of a digital elevation model and one using data from the ICESat GLAS laser altimeter. This study shows the power of satellite remote sensing for long‐term monitoring of regional‐scale hydrologic transformations.