变分法遥感影像人工地物自动检测

人工地物（建筑物、道路、桥梁等）检测是目标识别的一个重要组成部分。本文将人工地物检测转换为能量泛函数最优化问题。首先对遥感影像进行超像素分割,综合图像的颜色、纹理、梯度等信息,以超像素为单元计算图像的显著度信息,然后构建一个包含显著性约束、面积和边界约束、纹理约束及灰度方差约束的能量泛函数,通过变分法迭代求解能量泛函最小值,获取目标前景部分即为人工地物区域。本文以重庆和广东某地的遥感影像数据为例对算法进行验证,将其与常见的人工地物目标提取算法,如C-V模型、MRF模型,以及当下研究较为热门的深度学习算法进行对比。试验结果表明,该算法能有效地检测出遥感影像中的人工地物区域,并保证较低的误检率及漏检率。论文对该方法与深度学习方法进行了一定的分析对比。     

深度残差网络的多光谱遥感图像显著目标检测

本文侧重于介绍智能化摄影测量深度学习的深度残差方法。显著目标检测致力于自动检测和定位图像中人最感兴趣的目标区域。多波段遥感图像因其更加丰富的光谱信息和揭示观测目标物理属性的能力在目标检测中获得重要应用。传统的显著目标检测方法通过手工设计特征,计算图像各像素或者超像素与邻域像素或者超像素之间的对比度检测显著目标。随着深度学习的巨大发展,特别是全卷积神经网络的引入,基于深度卷积网络的显著目标检测算法取得重要进步。然而,由于数据获取和标记的困难,多波段遥感图像显著目标检测的研究依然主要采用手工设计特征。本文研究基于深度卷积神经网络的多波段遥感图像显著目标检测算法,提出一种基于深度残差网络的自上而下的多光波段遥感图像显著目标检测网络,该网络可以有效挖掘深度残差网络不同层次上的显著性特征,以端对端方式实现显著目标检测。为了应对多波段遥感图像数据量有限、无法训练深度残差网络的问题,本文提出通过浅层神经网络从RGB图像直接生成多波段遥感图像,实现光谱方向的超分辨率。在现有多波段遥感图像和可见光图像显著目标检测数据集上的试验结果超过当前最好方法10%以上,验证了本文方法的有效性。     

Change detection based on DSM and image features in urban areas

On the basis of atereo image analysis,the change detection of man-made objects in urban areas is in-troduced. Information of the height of man-made objects can be applied to reinforce their change detection. By comparison between the new and old DSMs, the changed regions are extracted. However, our aim is to detect changes of man-made objects in urban area and further in the potental areas by the means of line-feature matching and gradient direction histogram. The experiments based on the aerial images from Japan have proven that the algorithm is correct and efficient.     

Region of interest extraction via common salient feature analysis and feedback reinforcement strategy for remote sensing images

Saliency detection is an effective approach to extract regions of interest (ROIs) for remote sensing images. However, existing saliency detection models mainly focus on ROI extraction from a single image and usually are not able to produce satisfactory results because of complex background interference in remote sensing images. The employment of mutual information in a set of remote sensing images can provide an effective solution to this issue. In this paper, we propose a novel saliency detection model for multiple remote sensing images to simultaneously extract ROIs and identify images without ROIs. First, common salient feature analysis based on synthesized feature clustering and global contrast is implemented to exploit global correspondence in the synthesis feature domain, thereby highlighting preliminary ROIs against background interference and assigning lower saliency values to images without ROIs. Then, we design an exclusion criterion based on saliency value judgment to remove images without ROIs, and the remaining saliency maps are refined by an enhancement strategy. Finally, the enhanced maps serve as a feedback to yield a homogenous synthesized feature space in which integral ROIs with subtle borders are extracted by the reused cluster-based saliency calculation. Experiments reveal that our model outperforms seven state-of-the-art models by achieving the best ROC curve (AUC = 0.945) and maximal F-measure at 0.729.     

Preliminary analytical method for unsupervised remote sensing image classification based on visual perception and a force field

Abstract

The analysis of remote sensing (RS) images, which is often accomplished using unsupervised image classification techniques, requires an effective method to determine an appropriate number of classification clusters. This paper proposes a preliminary analytical method to evaluate the input parameters for unsupervised RS image classification. Our approach involves first analysing the colour spaces of RS images based on the human visual perception theory. This enables the initial number of clusters and their corresponding centres to be automatically established based on the interaction of different forces in our supposed force field. The proposed approach can automatically determine the appropriate initial number of clusters and their corresponding centres for unsupervised image classification. A comparison of the experimental results with those of existing methods showed that the proposed method can considerably facilitate unsupervised image classification for acquiring accurate results efficiently and effectively without any prior knowledge.     

结合卷积神经网络与显著性特征的机场检测

遥感影像机场检测中，针对传统人工设计特征的方法稳健性差、检测耗时的问题，提出了一种结合卷积神经网络与显著性特征的遥感影像机场检测算法。利用卷积神经网络快速准确地检测出机场目标，确定兴趣区域，对兴趣区域进行显著性检测和连通区提取，从而获取更加精确的机场边界，最后利用多种场景下的影像进行测试。结果表明，本文方法具有明显的精度和速度优势；利用频率视觉显著性分析方法对获得的机场区域进行视觉显著性检测，可有效获取机场和跑道的精确边界，提高机场检测的效果和实用价值。     

Efficient,simultaneous detection of multi-class geospatial targets based on visual saliency modeling and discriminative learning of sparse coding

Automatic detection of geospatial targets in cluttered scenes is a profound challenge in the field of aerial and satellite image analysis. In this paper, we propose a novel practical framework enabling efficient and simultaneous detection of multi-class geospatial targets in remote sensing images (RSI) by the integration of visual saliency modeling and the discriminative learning of sparse coding. At first, a computational saliency prediction model is built via learning a direct mapping from a variety of visual features to a ground truth set of salient objects in geospatial images manually annotated by experts. The output of this model can predict a small set of target candidate areas. Afterwards, in contrast with typical models that are trained independently for each class of targets, we train a multi-class object detector that can simultaneously localize multiple targets from multiple classes by using discriminative sparse coding. The Fisher discrimination criterion is incorporated into the learning of a dictionary, which leads to a set of discriminative sparse coding coefficients having small within-class scatter and big between-class scatter. Multi-class classification can be therefore achieved by the reconstruction error and discriminative coding coefficients. Finally, the trained multi-class object detector is applied to those target candidate areas instead of the entire image in order to classify them into various categories of target, which can significantly reduce the cost of traditional exhaustive search. Comprehensive evaluations on a satellite RSI database and comparisons with a number of state-of-the-art approaches demonstrate the effectiveness and efficiency of the proposed work.     

Analysis of large-scale UAV images using a multi-scale hierarchical representation

Abstract

Unmanned aerial vehicle (UAV)-based imaging systems have many superiorities compared with other platforms, such as high flexibility and low cost in collecting images, providing wide application prospects. However, the acquisition of the UAV-based image commonly results in very high resolution and very large-scale images, which poses great challenges for subsequent applications. Therefore, an efficient representation of large-scale UAV images is necessary for the extraction of the required information in a reasonable time. In this work, we proposed a multi-scale hierarchical representation, i.e. binary partition tree, for analyzing large-scale UAV images. More precisely, we first obtained an initial partition of images by an oversegmentation algorithm, i.e. the simple linear iterative clustering. Next, we merged the similar superpixels to build an object-based hierarchical structure by fully considering the spectral and spatial information of the superpixels and their topological relationships. Moreover, objects of interest and optimal segmentation were obtained using object-based analysis methods with the hierarchical structure. Experimental results on processing the post-seismic UAV images of the 2013 Ya’an earthquake and the mosaic of images in the South-west of Munich demonstrate the effectiveness and efficiency of our proposed method.     

利用视觉注意模型和局部特征的遥感影像检索方法

利用尺度不变特征变换(scale invariant feature transform,SlFT)算子直接提取遥感影像局邵特征进行检索时存在关键点数目多、特征维数高等问题,因此,本文利用视觉注意模型,根据目标显著性的大小从影像上提取显著目标区域,并采用K-means聚类方法对提取的SIF丁局邵特征进行聚类,得到用于检索的特征向量实验结果表明,该方法不仅符合人眼的视觉特性,且在降低SI FT关键点数目和特征维数的同时提高了检索精度和检索效率.     

UAV navigation system using line-based sensor pose estimation

Abstract

This work presents a mapping and tracking system based on images to enable a small Unmanned Aerial Vehicle (UAV) to accurately navigate in indoor and GPS-denied outdoor environments. A method is proposed to estimate the UAV’s pose (i.e., the 3D position and orientation of the camera sensor) in real-time using only the on-board RGB camera as the UAV travels through a known 3D environment (i.e., a 3D CAD model). Linear features are extracted and automatically matched between images collected by the UAV’s onboard RGB camera and the 3D object model. The matched lines from the 3D model serve as ground control to estimate the camera pose in real-time via line-based space resection. The results demonstrate that the proposed model-based pose estimation algorithm provides sub-meter positioning accuracies in both indoor and outdoor environments. It is also that shown the proposed method can provide sparse updates to correct the drift from complementary simultaneous localization and mapping (SLAM)-derived pose estimates.     

结合数据融合与特征选择的遥感影像尺度多样目标检测

基于深度神经网络模型的遥感影像地物检测取得了巨大成功,很大程度上得益于大规模数据集的支撑。但是,从现有遥感影像数据集本身来看,不同类别地物的数量分布不一致,同类地物对象以不同尺寸大小呈现,是导致地物样本的尺度不均衡问题的直接因素。对此,本文采用数据集内影像加权融合与地物多尺度特征选择的策略来缓解该问题。首先,将数据集内两张影像的像素值进行加权并得到融合后的影像,从而使不同类别地物样本更加均衡且具有较高的背景多样性;其次,通过选择合适尺度的特征图预测相应尺度的目标类别,且允许同一尺度目标在相邻特征图上进行预测,这样使模型能根据目标尺度进行训练;最后,基于目标中心区域的特征图预测目标边界框,预测的边界框更符合目标本身的尺度。通过在两个遥感数据集上分别进行实验,表明训练的模型在对复杂背景下的类别不均衡目标的识别更加准确,能够适应遥感影像下不同尺度目标的识别。     

GVF Snake与显著特征相结合的高分辨率遥感图像道路提取

高分辨率遥感图像中的道路信息,对地理信息系统数据库的更新具有重要的意义。本文通过分析道路在遥感图像中所呈现的特性,提出了一种基于显著特征和GVF Snake的高分辨率遥感图像道路提取方法。该方法根据视觉认知理论将道路的几何特性和辐射特征作为显著特性。首先,通过融合颜色对比度和空间统计特征计算显著性图,并以输出的显著图的最大值作为GVF Snake的初始种子点;再利用区域生长法求出道路的初始边界,通过梯度矢量流模型的迭代求解,并最小化能量函数,实现道路信息的自动提取。试验结果表明,本文所提出的方法提高了不仅可以提高计算效率,还具有较好的检测精度。     

低空UAV激光点云和序列影像的自动配准方法

提出了一种低空无人机(unmanned aerial vehicle,UAV)序列影像与激光点云自动配准的方法。首先分别基于多标记点过程与局部显著区域检测对激光点云和序列影像的建筑物顶部轮廓进行提取,并依据反投影临近性匹配提取的顶面特征。然后利用匹配的建筑物角点对,线性解算序列影像外方位元素,再使用建筑物边线对的共面条件进行条件平差获得优化解。最后,为消除错误提取与匹配特征对整体配准结果的影响,使用多视立体密集匹配点集与激光点集进行带相对运动阈值约束的ICP(迭代最临近点)计算,整体优化序列影像外方位元素解。试验结果表明本文方法能实现低空序列影像与激光点云像素级精度的自动配准,联合制作DOM精度满足现行无人机产品1∶500比例尺标准。     

Urban accessibility diagnosis from mobile laser scanning data

In this paper we present an approach for automatic analysis of urban accessibility using 3D point clouds. Our approach is based on range images and it consists in two main steps: urban objects segmentation and curbs detection. Both of them are required for accessibility diagnosis and itinerary planning.Our method automatically segments facades and urban objects using two hypotheses: facades are the highest vertical structures in the scene and objects are bumps on the ground on the range image. The segmentation result is used to build an urban obstacle map. After that, the gradient is computed on the ground range image. Curb candidates are selected using height and geodesic features. Then, nearby curbs are reconnected using Bézier curves. Finally, accessibility is defined based on geometrical features and accessibility standards.Our methodology is tested on two MLS databases from Paris (France) and Enschede (The Netherlands). Our experiments show that our method has good detection rates, is fast and presents few false alarms. Our method outperforms other works reported in the literature on the same databases.     

车载激光扫描数据中多类目标的层次化提取方法

提出了一种从车载激光扫描数据中层次化提取多类型目标的有效方法。该方法首先利用颜色、激光反射强度、空间距离等特征,生成多尺度超级体素;然后综合超级体素的颜色、激光反射强度、法向量、主方向等特征利用图分割方法对体素进行分割;同时计算分割区域的显著性,以当前显著性最大的区域为种子区域进行邻域聚类得到目标;最后结合聚类区域的几何特性判断目标可能所属的类别,并按照目标类别采用不同的聚类准则重新聚类得到最终目标。试验结果表明,该方法成功地提取出建筑物、地面、路灯、树木、电线杆、交通标志牌、汽车、围墙等多类目标,目标提取的总体精度为92.3%。     

无人机影像辅助下的路桥病害智能检测

路桥表面病害状况评估是路桥养护的一项重要内容。目前病害检测主要基于移动测量车和目视判断,具有工作量大、获取危险度高的缺点,而低空飞行的六旋翼无人机能够拍摄到人工无法获取到的多角度路桥照片,在路桥检测方面具有巨大的优势。本文基于无人机影像开展路桥病害检测相关研究,提出了一种新的路桥病害检测方法。首先通过多部件形变模型模拟病害目标,并在无人机影像中全局搜索,检测出潜在路桥病害区域。试验表明,本文算法在复杂背景下能够有效检测病害,目标检测精度达80%,具有很高的效率和鲁棒性。     

单张无人机影像水体高光自动检测与补偿方法

无人机获取地面影像时,水体反射会导致影像中存在较为明显的水体高光,给数据处理带来一定困难并对DOM（digital orthophoto map）质量产生显著影响。提出了一种单张无人机影像水体高光自动检测与补偿方法,先基于高光分量对高光区域进行多尺度检测,再使用Grabcut算法对其边界进行优化;然后根据地物反射的几何与光谱特性建立决策树剔除非水体高光;接着精化高光区域,将附近的细碎高光点纳入到区域内;最后,使用提出的改进Criminisi算法对高光区域进行补偿。使用无人机影像对该方法的有效性进行验证,综合主观目视与PNSR（peak signal to noise ratio）、SSIM（structural similarity index）指数的客观评价结果,本文方法优于Mallick、Shen、Yoon等人的方法;使用Pix4Dmapper对高光处理前后的影像生成DOM,本文方法可以明显提高DOM质量。     

遥感影像目标检测的轻量化密集连接网络北大核心CSCD

针对无人机遥感影像中可能出现的多个密集的小目标,在检测时会出现误检、漏检的难点问题,本文提出了一种基于YOLO v4的具有密集连接网络模块的遥感影像轻量化目标检测算法,实现了对无人机遥感影像中车辆小目标的高精度识别。首先,对YOLO v4主干网络CSP Darknet53的卷积层采用密集连接、稀疏连接两种处理方式,加强特征的提取和重复使用,以缓解梯度消失问题;然后,对此模块进行模型剪裁,减少网络层数并定义为新的密集连接网络模块;最后,在NWPU-VHR-10数据集和笔者所在课题组制作的Vehicle-850无人机影像数据集上进行了对比试验并取得了较好的效果。本文改进后的网络结构在提高遥感影像目标检测准确率的同时,缩短了网络模型的收敛时间,减少了模型占用的内存空间,提高了遥感影像中目标检测的速度。     

Accuracy assessment of real-time kinematics (RTK) measurements on unmanned aerial vehicles (UAV) for direct geo-referencing

ABSTRACT

Geospatial information acquired with Unmanned Aerial Vehicles (UAV) provides valuable decision-making support in many different domains, and technological advances coincide with a demand for ever more sophisticated data products. One consequence is a research and development focus on more accurately referenced images and derivatives, which has long been a weakness especially of low to medium cost UAV systems equipped with relatively inexpensive inertial measurement unit (IMU) and Global Navigation Satellite System (GNSS) receivers. This research evaluates the positional accuracy of the real-time kinematics (RTK) GNSS on the DJI Matrice 600 Pro, one of the first available and widely used UAVs with potentially surveying-grade performance. Although a very high positional accuracy of the drone itself of 2 to 3 cm is claimed by DJI, the actual accuracy of the drone RTK for positioning the images and for using it for mapping purposes without additional ground control is not known. To begin with, the actual GNSS RTK position of reference center (the physical point on the antenna) on the drone is not indicated, and uncertainty regarding this also exists among the professional user community. In this study the reference center was determined through a set of experiments using the dual frequency static Leica GNSS with RTK capability. The RTK positioning data from the drone were then used for direct georeferencing, and its results were evaluated. Test flights were carried out over a 70 x 70 m area with an altitude of 40 m above the ground, with a ground sampling distance of 1.3 cm. Evaluated against ground control points, the planimetric accuracy of direct georeferencing for the photogrammetric product ranged between 30 and 60 cm. Analysis of direct georeferencing results showed a time delay of up to 0.28 seconds between the drone GNSS RTK and camera image acquisition affecting direct georeferencing results.     

无人机采场图像二维经验小波变换降噪研究

针对传统图像降噪方法难以适用于露天矿无人机(UAV)图像降噪的问题,该文提出了一种基于经验小波变换(EWT)算法思想的露天矿UAV图像快速降噪方法。采用Littlewood-Paley小波算子检测提取UAV图像噪声特征,以提高阈值函数匹配精度;借助改进的自适应H阈值函数对UAV图像噪声进行计算,提高UAV图像边缘细节度;并通过二维经验小波变换逆重构,从而提高UAV图像纹理精度。实验结果表明,本文算法能够实现露天矿UAV图像的快速降噪,保留边缘细节和纹理特征,并具有较好的降噪效果。