基于超像元词包特征和主动学习的高分遥感影像变化检测     

杨进一  徐伟铭  王成军  翁谦 《地球信息科学学报》2019,21(10):1594-1607

为解决高分辨率遥感影像变化检测中存在底层特征缺乏语义信息、像元级的检测结果存在“椒盐”现象以及监督分类中样本标注自动化程度较低,本文提出一种基于超像元词包特征和主动学习的变化检测方法。首先采用熵率分割算法获取叠加影像的超像元对象;其次提取两期影像像元点对间的邻近相关影像特征（相关度、斜率和截距）和顾及邻域的纹理变化强度特征（均值、方差、同质性和相异性）,经线性组合作为像元点对的底层特征;然后基于像元点对底层特征利用BOW模型构建超像元词包特征,并采用一种改进标注策略的主动学习方法从无标记样本池中优选信息量较大的样本,且自动标注样本类别;最后训练分类器模型完成变化检测。通过选用2组不同地区的GF-2影像和Worldview-Ⅱ影像作为数据源进行实验,实验结果中2组数据集的F1分数分别为0.8714、0.8554,正确率分别为0.9148、0.9022,漏检率分别为0.1681、0.1868,误检率分别为0.0852、0.0978。结果表明,该法能有效识别变化区域、提高变化检测精度。此外,传统主动学习方法与改进标注策略的主动学习方法的学习曲线对比显示,改进的标注策略可在较低精度损失下,有效提高样本标注自动化程度。  相似文献   

海面溢油无人机高光谱遥感检测与厚度估算方法   总被引：2，自引：0，他引：2  

任广波  过杰  马毅  罗旭东 《海洋学报》2019,41(5):146-158

海上溢油是海洋国家所面临的共同问题，但至今仍没有一种可靠实用的海上溢油准确识别和油量遥感监测方法。为此，本文以无人机高光谱遥感为手段，开展了海面溢油检测与厚度估算方法研究。实验中，通过搭建室外大型水槽溢油实验装置，获取了模拟真实海洋环境条件下不同溢油量的遥感和现场光谱数据，在此基础上，分析并提取了海上溢油特征光谱波段，给出了海上溢油高光谱检测模型；针对现场实验条件下水面油膜厚度难以测定的问题，设计了3种利用总体溢油量的油膜厚度估算模型。得到如下主要结论:（1）675 nm和699 nm是海上溢油检测的有效特征波段，但对极薄的油膜没有检测能力；（2）提出了归一化溢油指数模型、反比例模型和吸收基线模型等3种海上溢油油膜厚度估算模型，其中对于薄油膜（厚度≤ 5 μm）和厚油膜（厚度>50 μm），反比例模型是溢油厚度反演的首选也是唯一选择。对于中厚度油膜，晴朗天气条件下，归一化溢油指数模型是油膜厚度反演的首选，同时反比例模型和溢油吸收基线模型也都有较好的反演能力，而在多云天气条件下，反比例模型效果最佳。  相似文献   

顾及地形起伏的无人机影像覆盖分析方法     

张衡  李骁  叶朋飞 《测绘通报》2020,(1):102-106

使用无人机实施测绘航空摄影时，由于无人机相对航高较低，地面起伏会对无人机影像的分辨率、覆盖范围、重叠度造成较大的影响，影像成果会出现分辨率不足、重叠度不够、覆盖漏洞等缺陷。针对这一情况，本文提出了一种利用数字微分正解法的计算方法，借助DEM准确计算每张影像的覆盖范围，并使用FME软件高效生成全部影像的覆盖范围。经过实际使用，验证了该方法可以在航线设计阶段准确预测并分析全部影像的覆盖范围、重叠度，因此可及时发现设计问题并调整航线。该方法可以有效减少因地形起伏造成的影像覆盖缺陷，减少返工现象，从整体上提高了作业效率。  相似文献   

低空无人机的倾斜摄影测量技术在城市轨道交通建设中的应用     

冯增文  李珂  李梓豪  任传斌 《测绘通报》2020,(9):42-45

智慧地铁的建设需要三维模型的支撑，运用倾斜摄影测量技术，结合三维可视化技术，为智慧地铁提供了三维数据支撑，推动了全生命周期智慧地铁建设的全面发展。本文旨在以雄安新区城市轨道交通R1线为研究区，对其沿线85 km进行了无人机航飞工作，通过数据处理获取了沿线三维实景模型，并对模型成果精度进行了详细对比。结果表明，此次倾斜摄影测量成果满足1：500比例尺要求。此外，还对倾斜摄影测量技术在城市轨道交通建设中的相关应用进行了介绍，为城市轨道交通建设工作由二维向三维转变开拓了新的方向。  相似文献   

针对地质灾害隐患点的无人机航摄像控点布点方案     

吴波  李俊  蒋大鹏  冯志 《测绘通报》2020,(6):125

以陕西省某地质灾害隐患点1∶2000航飞生产任务为背景，本文结合垂直起降式无人机航摄技术特点和地质灾害隐患点区域实际情况，通过详细对比几种无人机航摄像控点布控方案的优劣，研究无人机像控点布设方案对空三加密精度的影响，总结了适用于不同地质灾害隐患点的像控点布设方案。通过本文研究，为今后制定地质灾害隐患点的无人机航摄像控点布设方案提供了参考依据。  相似文献   

基于深度学习理论的无人机影像分类     

阳成 《北京测绘》2020,(4):481-484

针对无人机影像深度学习分类方法缺乏现状,本文利用深度学习理论卷积神经网络方法对无人机影像进行了分类。该法首先抽取无人机影像作为训练集和检验集,然后建立一个2个卷积层-池化层的卷积神经网络模型进行深度学习,通过设定参数并运行模型实现无人机影像分类。实验表明,本文提出的方法可完成较复杂地区无人机影像分类,其分类精度与支持向量机方法相当,为无人机遥感影像分类提供了一个崭新的技术视点。  相似文献   

Monitoring soil surface roughness under growing winter wheat with low-altitude UAV sensing: Potential and limitations     

Nils Onnen  Anette Eltner  Goswin Heckrath  Kristof Van Oost 《地球表面变化过程与地形》2020,45(14):3747-3759

Soil surface roughness (SSR) is an important factor in controlling sediment and runoff generation, influencing directly a wide spectrum of erosion parameters. SSR is highly variable in time and space under natural conditions, and characterizing SSR to improve the parameterization of hydrological and erosion models has proved challenging. Our study uses recent technological and algorithmic developments in capturing and processing close aerial sensing data to evaluate how high-resolution imagery can assist the temporally and spatially explicit monitoring of SSR. We evaluated the evolution of SSR under natural rainfall and growing vegetation conditions on two arable fields in Denmark. Unmanned aerial vehicle (UAV) photogrammetry was used to monitor small field plots over 7 months after seeding of winter wheat following conventional and reduced tillage treatments. Field campaigns were conducted at least once a month from October until April, resulting in nine time steps of data acquisition. Structure from motion photogrammetry was used to derive high-resolution point clouds with an average ground sampling distance of 2.7 mm and a mean ground control point accuracy of 1.8 mm. A comprehensive workflow was developed to process the point clouds, including the detection of vegetation and the removal of vegetation-induced point cloud noise. Rasterized and filtered point clouds were then used to determine SSR geostatistically as the standard deviation of height, applying different kernel sizes and using semivariograms. The results showed an influence of kernel size on roughness, with a value range of 0.2–1 cm of average height deviation during the monitoring period. Semivariograms showed a measurable decrease in sill variance and an increase in range over time. This research demonstrated multiple challenges to measuring SSR with UAV under natural conditions with increasing vegetation cover. The proposed workflow represents a step forward in tackling those challenges and provides a knowledge base for future research. © 2020 John Wiley & Sons, Ltd.  相似文献   

Evaluating functional connectivity in a small agricultural catchment under contrasting flood events by using UAV     

Aleix Calsamiglia  Jorge Gago  Julián Garcia-Comendador  Josep Fortesa Bernat  Adolfo Calvo-Cases  Joan Estrany 《地球表面变化过程与地形》2020,45(4):800-815

Concentrated erosion, a major feature of land degradation, represents a serious problem for soil and water resources management and a threat to ecosystems. Understanding the internal mechanisms (de-)coupling sediment pathways can improve the management and resilience of catchments. In this study, concentrated erosion and deposition forms were mapped accurately through field and aerial unmanned aerial vehicle (UAV) campaigns, in order to assess the evolution of connectivity pathways over a series of three contrasted and consecutive flood events occurring between October 2016 and January 2017 (return period ranging from 0.5 to 25 years) in a small Mediterranean agricultural catchment (Can Revull, Mallorca, Spain; 1.4 km²). In addition, a morphometric index of connectivity (IC) was used to identify the potential trajectories of different concentrated erosion forms and deposition areas. IC predictions were calibrated by identifying the optimal critical thresholds, i.e. those most consistent with field observations after each of the events studied. The results found that the index performed well in predicting the occurrence and the length/area of the different type of landforms, giving kappa (κ) coefficients of variation ranging between 0.21 and 0.92 and linear correlations R² between 0.33 and 0.72. The type of landform affected the correspondence of IC predictions and field observations, with lower thresholds the greater the magnitude of their associated geomorphic processes. Rainfall magnitude proved to be a very important factor controlling the development of erosion and deposition landforms, with large differences in length/area between the contrasted events. The evolution of the observed trajectories revealed feedback dynamics between the structural and functional connectivity of the catchment, in which morphological changes determined the spatial distribution of the processes’ activity in the successive events and vice versa. © 2020 John Wiley & Sons, Ltd.  相似文献   

无人机在露天矿山监测中的应用   总被引：3，自引：0，他引：3  

白洋  康会涛  张文超  贾玉娜  郑浩 《测绘通报》2020,(9):85-88

在露天矿山开采过程中，对矿山进行实时动态监测很有必要，它不仅能够实时掌握开采进度和开采方案实施的准确性，还有利于数字矿山建设。本文以河北省保定市涞源县某露天矿区为例，利用Pix4Dmapper软件对无人机航拍影像进行数据采集与处理，获取矿区的三维点云数据、数字正射影像和数字地表模型，并利用ArcGIS和Global Mapper软件对生产成果进行矿山开发占地分析，实现露天矿山三维可视化。本文为准确掌握矿山环境治理执行情况，把握矿山开发整体情况给予了有力的参考和借鉴作用。  相似文献   

Inter-comparison of remote sensing platforms for height estimation of mango and avocado tree crowns     

《International Journal of Applied Earth Observation and Geoinformation》2020

To support the adoption of precision agricultural practices in horticultural tree crops, prior research has investigated the relationship between crop vigour (height, canopy density, health) as measured by remote sensing technologies, to fruit quality, yield and pruning requirements. However, few studies have compared the accuracy of different remote sensing technologies for the estimation of tree height. In this study, we evaluated the accuracy, flexibility, aerial coverage and limitations of five techniques to measure the height of two types of horticultural tree crops, mango and avocado trees. Canopy height estimates from Terrestrial Laser Scanning (TLS) were used as a reference dataset against height estimates from Airborne Laser Scanning (ALS) data, WorldView-3 (WV-3) stereo imagery, Unmanned Aerial Vehicle (UAV) based RGB and multi-spectral imagery, and field measurements. Overall, imagery obtained from the UAV platform were found to provide tree height measurement comparable to that from the TLS (R² = 0.89, RMSE = 0.19 m and rRMSE = 5.37 % for mango trees; R² = 0.81, RMSE = 0.42 m and rRMSE = 4.75 % for avocado trees), although coverage area is limited to 1–10 km² due to battery life and line-of-sight flight regulations. The ALS data also achieved reasonable accuracy for both mango and avocado trees (R² = 0.67, RMSE = 0.24 m and rRMSE = 7.39 % for mango trees; R² = 0.63, RMSE = 0.43 m and rRMSE = 5.04 % for avocado trees), providing both optimal point density and flight altitude, and therefore offers an effective platform for large areas (10 km²–100 km²). However, cost and availability of ALS data is a consideration. WV-3 stereo imagery produced the lowest accuracies for both tree crops (R² = 0.50, RMSE = 0.84 m and rRMSE = 32.64 % for mango trees; R² = 0.45, RMSE = 0.74 m and rRMSE = 8.51 % for avocado trees) when compared to other remote sensing platforms, but may still present a viable option due to cost and commercial availability when large area coverage is required. This research provides industries and growers with valuable information on how to select the most appropriate approach and the optimal parameters for each remote sensing platform to assess canopy height for mango and avocado trees.  相似文献