基于迁移学习的低空摄影测量滑坡方量估算方法

针对现有方法难以准确地估算山体滑坡体积的问题,引入人工智能算法,提出耦合迁移学习与微分算法的低空摄影测量山体滑坡方量估算方法。首先,利用SfM与SGM密集匹配等算法从低空无人机立体影像中解算出高精度三维密集点云,结合可见光植被指数和双边滤波算法从密集点云中剥离出目标区地面点云;然后,构建深度神经网络插值模型来表征二维坐标与高程之间的非线性映射关系,并基于参数共享的迁移学习来自适应优化深度神经网络以实现滑坡目标区高程值预测,进而重构滑坡区域的数字地表模型;最后,基于目标区滑坡前后数字地表模型高程差值和微分算法实现山体滑坡方量估算。实验结果表明,该方法平均相对误差为2.7%,相比常用的方法,显著提高了滑坡方量估计精度,并能适应不同地形条件下滑坡方量估算。     

基于高精度低空摄影测量的滑坡灾害监测与预报关键技术研究

采用传统监测与预报技术,受到滑坡区域扩大影响,导致预测精准度较低,针对该问题,提出了基于高精度低空摄影测量的滑坡灾害监测与预报关键技术研究。依据滑坡地质体环境因素,建立滑坡灾害监测与监测与预报指标,根据该指标,计算地层压力变量相关参数。以高分辨率数字遥感设备为传感器,采用高精度低空摄影测量技术来提取数据,以此分析滑坡堆积特征和密实程度。通过上述特性,建立关联函数,确定监测与预报对象以及监测与预报等级,由此完成滑坡灾害监测与预报。由实验结果可知,该技术预测精准度较高,最高可达到98.8%,为保障人们生命安全提供必要技术支持。     

ENVI SARscape高级雷达数据处理工具

合成孔径雷达（SAR）拥有独特的技术魅力和优势,逐渐成为国际上的研究热点之一,其应用领域越来越广泛。SAR数据可以全天候对研究区域进行量测、分析以及获取目标信息。InSAR技术可提取地形信息和地表形变信息,SAR及其干涉测量技术主要应用于地形数据（DEM）提取、地表沉降监测、滑坡／冰川移动监测、目标识别与跟踪、原油泄漏...     

基于WorldView-2制备大野口流域高分辨率DEM及精度分析

在全野外GPS地面控制点基础上,对WorldView-2影像自带RPC文件进行校正,利用数字摄影测量软件系统在立体模型上通过影像自动匹配技术快速提取黑河流域上游大野口子流域1∶5 000比例尺数字高程模型(DEM)。由于区域地形复杂、交通不便,研究区南部无地面控制点覆盖。基于立体模型交互式操作,匹配60个均匀分布高精度影像连接点,提高了DEM自动提取精度。并在对阴坡森林覆盖区、大野口水库等重点区域进行DEM编辑基础上,辅助地形特征点和线数据提高了成果精度。由15个外业控制点、12个模型保密点组成的检查点进行定量DEM验证,结果表明:两组高程中误差最大为1.9 m,达到该比例尺山地一级精度2.5 m的要求。
     

数字近景摄影测量技术在矿山地表沉陷监测中的应用研究

为了解决传统的矿山地表沉陷监测方法存在的劳动强度大、不能实时观测、难以获得瞬时3维移动变形信息等缺陷,提出了一种利用基于非量测数相机的数字近景摄影测量技术进行塌陷区沉陷监测的方法。该方法首先利用标定过的高分辨率数字相机获取塌陷区的数字立体影像像对;然后采用相对定向、影像匹配等摄影测量解析处理技术提取塌陷区地表的数字高程模型;最后通过与该区域开采前数字高程模型对比分析来完成塌陷区沉陷范围、深度、体积等沉陷参数计算。应用研究结果表明,数字近景摄影测量的测定精度基本可以满足这次应用条件下沉陷监测的精度要求。该方法为矿山地表塌陷区的沉陷监测提供了一条可行的途径。     

卷积神经网络支持下的低空摄影测量DEM修补

针对低空摄影测量DEM生成大多需要人工后处理存在自动化程度较低的问题,结合深度学习算法,提出卷积神经网络支持下的低空摄影测量DEM自动修补方法。该方法构建卷积神经网络低空遥感分类模型识别DEM修补目标区,采用高差能量衰减函数寻找最优的关联高程数据集,利用抗差径向神经网络高程曲面拟合法修补目标区高程值,以实现低空摄影测量DEM自动修补。实验验证了该方法可达到人工后处理DEM的精度,且显著提高了DEM修补的自动化程度。     

面向对象滑坡信息提取中DEM空间分辨率影响分析

数字高程模型(DEM)是识别滑坡的重要特征,而在实际应用中通常难以获取研究区的高分辨率DEM数据。为分析DEM分辨率对滑坡提取的影响并确定满足应用的分辨率要求,将原始10 m分辨率DEM重采样为空间分辨率不同的15、20、30、50、75与100 m,采用DEM及其衍生数据辅助高分遥感影像的面向对象滑坡信息提取方法进行实验。实验结果表明:当DEM分辨率小于30 m时,对面积大于5 000 m²的滑坡能得到较好的识别和分类;大于30 m时虽难以区分出滑坡类型,但通过调整部分规则参数仍能实现对滑坡的监测。该研究对于准确提取滑坡信息所需DEM产品的选择具有一定的指导意义和参考价值。     

ENVI SARscape高级雷达数据处理工具

合成孔径雷达（SAR）拥有独特的技术魅力和优势,渐成为国际上的研究热点之一,其应用领域越来越广泛。SAR数据可以全天候对研究区域进行量测、分析以及获取目标信息。InSAR技术可提取地形信息和地表形变信息,SAR及其干涉测量技术主要应用于地形数据（DEM）提取、地表沉降监测、滑坡／冰川移动监测、目标识别与跟踪、原油泄漏跟踪、作物生长跟踪、农作物产量评估、森林制图,以及洪水、火灾和地震的灾害评估等领域。     

无人机倾斜摄影在滑坡区域监测中的应用研究

为了提高滑坡区域监测能力，采用无人机倾斜摄影技术实现滑坡区域监测，同时采用特征点提取和匹配特征点的方法对滑坡监测区域数据信息特征进行监测。通过影像分析和变换，以提取所需特征的方法，提高数据信息应用和监测能力。将不同的硬件结构划分为电源模块和处理器、模拟/数字(A/D)转换模块、脉冲宽度调制(PWM)驱动模块、电平转换模块。通过这四个模块的配合，能够实现远程摄影相关工作。试验结果表明，该研究所运用的方法监测误差更小，最低达到4%以下。试验结果验证了方法的稳定性。     

大区域山体滑坡危险区域挖掘方法研究与仿真

研究山体滑坡的准确预警问题。山体滑坡形成过程极为复杂,发生时间较短,山体不同区域承压也有较大差别,各地区压力动态关联性不一致。传统的预测手段多是基于小区域,压力集中的条件下进行预测,一旦各地区承压关联出现混乱,会导致传统的单一山体承压模型反应滞后、甚至错误,造成挖掘不准。为了避免上述缺陷,提出了一种地层压力数据挖掘算法的山体滑坡预警方法。通过地层压力变量关系,计算地层压力变量相关参数,通过地层压力数据挖掘方法,对山体滑坡灾害进行预警。实验结果表明,提出的算法能够有效提高山体滑坡预警的准确性。     

Chinese satellite photogrammetry without ground control points based on a public DEM using an efficient and robust DEM matching method

Photogrammetry based on high-resolution satellite image can acquire geospatial information within a large area rapidly and timely, but its geopositioning accuracy is highly dependent on ground control points. Under the background of global mapping, a public digital elevation model (DEM) assisted Chinese satellite image geopositioning scheme was proposed to realize satellite photogrammetry without ground control points. To make full use of public DEM advantages of consistent and high accuracy, public DEM was regarded as reference data and matched with the DEM extracted from image, and then the determined transformation parameters were applied to correct direct georeferencing results in object space. A fast least Z-difference method combined with least trimmed squares estimator was proposed to achieve DEM matching, which can not only automatically select corresponding point determination model, but also self-adaptively detect and eliminate difference between DEMs. Multi-groups of comparative experiments using Mapping Satellite-1 and ZiYuan-3 surveying satellite images were designed. Experimental results show that the DEM assisted geopositioning scheme exploits advantages of reference DEM, which can greatly improve the accuracy of photogrammetry without ground control points to a relatively high level. The geopositioning accuracy of image is largely determined by, but not confined to, the accuracy of reference DEM, but it is slightly affected by the resolution of reference DEM. If target DEM resolution is relatively high, the geopositioning accuracy of a single image assisted by Shuttle Radar Topography Mission (SRTM) DEM can satisfy the accuracy requirements of 1:50,000 scale mapping perfectly. This scheme also has good robustness and high computational efficiency.     

Validation and comparison of different techniques for the derivation of digital elevation models and volcanic monitoring (Vulcano Island,Italy)

High accurate digital elevation models (DEM) acquired periodically over a volcanic area can be used for monitoring crustal deformations. Airborne stereoscopic photography is a powerful tool for the derivation of high resolution DEM, especially when combined with Global Positioning System (GPS). We analyse data acquired on Vulcano Island (Italy) to assess the performance of two photogrammetry methods for DEM generation. The first method is based on automatic digital processing of scanned airborne stereo images from a film camera (Wild RC20). In the second method digital stereo data from the multi-spectral High Resolution Stereo Camera-Airborne (HRSC-A) are used. Accuracy assessment through comparison with kinematic GPS height profiles shows that both DEMs have accuracy on the order of few decimetres. Direct comparison of the two DEMs on the La Fossa volcanic cone provides a standard deviation of the residuals of 78 cm. Residuals greater than two metres between the two DEMs acquired at one year interval are locally evidenced in unstable areas with uneven morphology. The application of photogrammetric DEMs is also discussed within a SAR interferometry study carried out on Vulcano Island to evaluate the potentialities of such techniques for ground deformation monitoring. Although accuracy better than 1 m or 2 m is not required for satellite SAR interferometry, we show how the precise photogrammetric DEMs could still significantly improve SAR interferograms of Vulcano Island.     

无人机遥感技术在河湖岸线监管中的应用

为提高河湖管理效率,针对无人机遥感技术观测范围广、获取空间信息速度快等优点,在分析无人机遥感技术工作原理基础上,根据河湖水域岸线监管不同需求,从目前无人机的多种监测方式入手,结合岸线监管应用,分析适用于河湖监管的 3 种监测方式。具体应用结果表明：1)低空摄影技术适合于小面积河湖岸线日常巡察,获取航拍照片。2)常规摄影测量技术是将外业获取的影像、POS、控制点数据,采用 PhotoScan 软件生成高分辨率正射影像和数字表面模型,结合 GIS 软件,快速生成一些专题图,适合于中小型涉水侵占行为监测。3)倾斜摄影测量技术适合于大型涉水建筑监测,对涉水建筑构建真三维模型,精准获取违建区域面积、侵占防洪库容等几何信息。无人机遥感技术用于河湖岸线监测,能够有效制止侵占岸线的行为,为河湖岸线安澜提供技术支撑。     

LIDAR点云支持的CCD影像地理编码

针对传统航空摄影测量生成有地理编码的正射影像需要一定的地面控制点、密集的同名点匹配和繁琐的DEM人工编辑,本文将LIDAR点云作为控制基准,采用共线、共面等几何约束条件,联合平差求出CCD影像的外方位元素,并采用LIDAR点云滤波后的DEM,对CCD影像进行正射纠正,生成具有地理编码的影像数据,可有效地提高工作效率,节省大量时间和人工成本,试验结果证实了这种思路和算法的可行性。     

无控制DEM匹配与地表差异探测

泥石流爆发地区控制点布设和维护困难,使得精确探测该类区域泥石流造成的地表变化非常困难。基于规则格网DEM（Digital Elevation Model）,分析泥石流灾害的特点,提出了顾及地形特征的无控制DEM匹配与差异探测方法。根据不同时相DEM数据,利用该方法成功探测出普歪沟泥石流爆发区的地表水土流失变化。     

Combining Structure from Motion and close-range stereo photogrammetry to obtain scaled gravel bar DEMs

ABSTRACT

Digital elevation models (DEMs) have been increasingly applied in topographic studies in areas such as physical geography and hydraulic engineering. Several methods have been proposed to reconstruct DEMs, including classic close-range stereo photogrammetry and the more novel Structure from Motion (SfM) methodology. Past published studies tend to apply SfM to large-scale environmental processes, whilst classic close-range stereophotogrammetry is focusing on detailed small-scale applications. However, SfM requires multiple ground control points (GCPs) to allow for proper DEM scaling. The larger the study area, the more GCPs are required, resulting in increased operational complexity and time-consuming application of SfM. As the accuracy of the DEM depends on the equipment used to measure GCPs, this can also result in a cost-expensive operation. In the present study, we introduce a combined SfM and close-range stereo photogrammetry application, with the close-range stereo photogrammetry results serving as a control for providing scale, thus eliminating the need for traditional GCPs. To validate our methodology, we studied a 40 m long gravel bar. We used GoPro Hero 3 cameras for SfM measurements and replaced GCPs by DEMs obtained through close-range stereo photogrammetry with a Nikon D5100 camera pair in stereo. In addition to using photo-mode frames, we also studied the quality of DEMs obtained with GoPro Hero 3 video-mode frames, and show how the DEM quality is reduced due to the smaller image format, hence coarser point cloud spacing, which eventually results in a convex curvature when image overlap was increased. Our results show that it is possible to collect high-quality topographic surface data by only using cameras, and alleviate the need for GCPs. The proposed workflow reduces the complexity, time, and resource demands associated with deploying GCPs and necessary independent geo-referencing, ensuring that digital photogrammetry will continue to gain popularity for field surveying.     

基于自适应末端滑膜控制的无人机倾斜摄影测量技术

传统的无人机倾斜摄影测量测量数据有效率低,得到的摄影图片清晰度差。为了解决上述问题,基于自适应末端滑膜控制研究了一种新的无人机倾斜摄影测量技术,利用垂直影像摄影,将数据进行初步收集追踪,并对倾斜摄影图像进行记录,结合空中三角测量方法,加强数据自身主导性,同时为无人机监测系统提供较好的数据来源,强化系统内部数据联合功能,在多视影像联合的过程中注意充分考虑影像间的遮挡关系及遮蔽条件,在经过对影像数据的联合收集后,对无人机倾斜摄影数据进行数据匹配操作,提升数据的整合性,并通过数据整理加强摄影影像的可操作性,利用无人机作为影像传输主要载体,进一步加强对倾斜影像的监测,确定影像空间位置,在数据集中的过程中加大滑膜控制,并经过控制点坐标及连接线方位确定辅助数据的具体位置。实验结果表明,基于自适应末端滑膜控制的无人机倾斜摄影测量技术能够得到更加清晰的摄影图片,测量数据的有效率提高了25.21%。