多尺度分割的高分辨率遥感影像变化检测

针对高空间分辨率的遥感影像,提出了一种基于多尺度分割的变化检测算法。采用Mean-Shift分割算法对影像进行多尺度分割,构建了不同尺度上的地理对象,以不同尺度上的地理对象灰度均值构建了变化检测的多尺度特征向量,采用变化矢量分析法获得最后的变化检测结果。以城镇区和农田区的Quick Bird影像对本文算法进行了检验,从精度评价的效果来看,无论城镇区还是农田区,采用面向对象的变化检测方法精度都高于基于单像素的检测方法,且当尺度层数固定时,多尺度组合的变化检测结果优于单一尺度的变化检测结果,对城镇、农田区域的变化检测的精度分别达到87.57%和81.55%。本文算法既可以顾及大面积同质区域变化,又可以反映小的地物目标及边缘部分的变化,能够很好地满足城镇、农田等不同环境背景下的变化检测需求,在国土资源监测中具有一定的应用价值。     

Automatic Registration Between Low-Altitude LiDAR Point Clouds and Aerial Images Using Road Features

Among the many means of acquiring surface information, low-altitude light detection and ranging (LiDAR) systems (e.g., unmanned aerial vehicle LiDAR, UAV-LiDAR) have become an important approach to accessing geospatial information. Considering the lower level of hardware technology in low-altitude LiDAR systems compared to that in airborne LiDAR, and the greater flexibility in-flight, registration procedures must be first performed to facilitate the fusion of laser point data and aerial images. The corner points and edges of buildings are frequently used for the automatic registration of aerial imagery with LiDAR data. Although aerial images and LiDAR data provide powerful support for building detection, adaptive edge detection for all types of building shapes is difficult. To deal with the weakness of building edge detection and reduce matching-related computation, the study presents a novel automatic registration method for aerial images, with LiDAR data, on the basis of main-road information in urban areas. Firstly, vector road centerlines are extracted from raw LiDAR data and then projected onto related aerial images with the use of coarse exterior orientation parameters (EOPs). Secondly, the corresponding image road features of each LiDAR vector road are determined using an improved total rectangle-matching approach. Finally, the endpoints of the conjugate road features obtained from the LiDAR data and aerial images are used as ground control points in space resection adjustment to refine the EOPs; an iterative strategy is used to obtain optimal matching results. Experimental results using road features verify the feasibility, robustness and accuracy of the proposed approach.     

面向区域与要素的城市1∶2000地形图数据库增量更新方法

大比例尺地形图数据库的更新是一项长期的重要任务。本文分析了1∶2000地形图数据库快速更新的难点,提出了一种航空摄影与卫星遥感、区域更新与要素更新相结合的大比例尺地形图数据库半自动快速更新方法。采用变化检测方法从更新前后卫星遥感影像中提取变化区域和变化要素,然后分别采用面向区域和面向要素的方法从高分辨率航空影像上测量变化地物,最后通过半自动空间实体匹配的方法建立现状库与历史库中要素的回溯关联,从而实现地形图数据库的半自动增量式快速更新。利用该方法对中山市东区的1∶2000地形图进行了更新试验。试验结果表明,引入卫星影像进行自动变化检测后,在航空影像上分区域和要素两种模式采集更新城市大比例尺地形图数据库,效率比传统方法提高25%。     

Tree Detection in Urban Regions Using Aerial Lidar and Image Data

In this letter, we present an approach to detecting trees in registered aerial image and range data obtained via lidar. The motivation for this problem comes from automated 3-D city modeling, in which such data are used to generate the models. Representing the trees in these models is problematic because the data are usually too sparsely sampled in tree regions to create an accurate 3-D model of the trees. Furthermore, including the tree data points interferes with the polygonization step of the building roof top models. Therefore, it is advantageous to detect and remove points that represent trees in both lidar and aerial imagery. In this letter, we propose a two-step method for tree detection consisting of segmentation followed by classification. The segmentation is done using a simple region-growing algorithm using weighted features from aerial image and lidar, such as height, texture map, height variation, and normal vector estimates. The weights for the features are determined using a learning method on random walks. The classification is done using the weighted support vector machines, allowing us to control the misclassification rate. The overall problem is formulated as a binary detection problem, and the results presented as receiver operating characteristic curves are shown to validate our approach     

Assessment of Shuttle Radar Topography Mission Elevation Data Based on Topographic Maps in Turkey

Depending on scale, topographic maps depicting the shape of the land surfaces of the Earth are produced from different data sources. National topographic maps at a scale of 1:25 000 (25K maps) produced by General Command of Mapping are used as the base map set in Turkey. This map set, which consists of approximately 5500 sheets, covers the whole country and is produced using photogrammetric methods. Digital Elevation Models (DEMs) created from these maps are also available. Recently, another data source, Synthetic Aperture Radar (SAR) interferometric data, has become more important than those produced by conventional methods. The Shuttle Radar Topography Mission (SRTM) contains elevation data with 3 arc-second resolution and 16 m absolute height error (90 percent confidence level). These data are freely available via the Internet for approximately 80 percent of the Earth's land mass. In this study, SRTM DEM was compared with DEM derived from 25K topographic maps for different parts of Turkey. The study areas, each covering four neighboring 25K maps, and having an area of approximately 600 km², were chosen to represent various terrain characteristics. For the comparison, DEMs created from the 25K maps were obtained and organized as files for each map sheet in vector format, containing the digitized contour lines. From these data, DEMs in the resolution of 3 arc-second were created (25K-DEM), in the same structure as the SRTM DEM, allowing the 25K-DEMs and the SRTM DEM to be compared directly. The results show that the agreement of SRTM DEM to the 25K-DEM is within about 13 m, which is less than the SRTM's targeted error of 16 m. The spatial distribution of the height differences between SRTM-DEM and the 25K-DEM and correlation analysis show that the differences were mainly related to the topography of the test areas. In some areas, local height shifts were determined.     

基于相位均匀卷积的LiDAR深度图与航空影像高效匹配方法

多源影像匹配主要受到非线性强度差异、对比度差异及局部区域结构特征不显著等问题的干扰,而机载激光雷达（light detection and ranging,LiDAR）深度图与航空影像由于纹理特征之间的显著差异导致部分结构特征缺失更为严重,所造成的相位极值突变进一步增加了匹配难度。因此,提出一种基于相位均匀卷积描述子的方法来实现LiDAR深度图与航空影像之间的高效匹配。在影像特征匹配阶段,首先对相位一致性模型进行扩展,构造相位均匀能量卷积方程,求解得到均匀卷积序列与相位最大标签图,建立一种相位均匀能量卷积直方图（histogram of phase mean energy convolution,HPMEC）;然后采用最近邻匹配算法完成初始匹配,并利用快速边缘化样本共识进行粗差剔除;最后基于线程池并行策略,通过划分重叠格网对影像进行分块匹配。将多组具有不同地物覆盖类型的LiDAR深度图与航空影像作为数据集,分别与位置尺度方向-尺度不变特征转换（position scale orientation-scale invariant feature transform,PSO-SIFT）、L...     

基于空-地多源数据融合的建筑物精细建模研究

针对建筑物精细建模的精度问题,对同一试验区建筑物群的机载LiDAR顶面数据、航空正射影像数据和车载LiDAR立面数据进行试验研究,通过精确提取各建筑物群的顶面和底面矢量轮廓线,对矢量轮廓线间的水平间距进行定性与定量分析,研究结果对空-地多源数据融合进行建筑物精细建模提供可靠的技术支持。     

Optimal fusion of optical and SAR high-resolution images for semiautomatic building detection

Building detection from different high-resolution aerial and satellite images has been a notable research topic in recent decades. The primary challenges are occlusions, shadows, different roof types, and similar spectral behavior of urban covers. Integration of different data sources is a solution to supplement the input feature space and improve the existing algorithms. Regarding the different nature and unique characteristics of optical and radar images, there are motivations for their fusion. This paper is aimed to identify an optimal fusion of radar and optical images to overcome their individual shortcomings and weaknesses. For this reason, panchromatic, multispectral, and radar images were first classified individually, and their strengths and weaknesses were evaluated. Different feature-level fusions of these data sets were then assessed followed by a decision-level fusion of their results. In both the feature and decision levels of integration, artificial neural networks were applied as the classifiers. Several post-processing methods using normalized different vegetation index, majority filter, and area filter were finally applied to the results. Overall accuracy of 92.8% and building detection accuracy of 89.1% confirmed the ability of the proposed fusion strategy of optical and radar images for building detection purposes.     

数字化航测调绘关键环节技术探讨

测绘技术的发展促使测绘作业方式发生巨大变化,传统的航测调绘方法逐渐被数字化的作业模式所取代。数字化航测调绘作为一种新兴的调绘方法,已经广泛应用于生产中,但是,在诸多关键环节上还有不少技术需要进一步研讨。本文针对该方法在原始资料的收集、分析,工作底图的制作,先内后外的工艺流程等诸多关键环节方面进行了细致的探讨。     

Object-based urban land cover classification using rule inheritance over very high-resolution multisensor and multitemporal data

Very high spatial and temporal resolution remote sensing data facilitate mapping highly complex and diverse urban environments. This study analyzed and demonstrated the usefulness of combined high-resolution aerial digital images and elevation data, and its processing using object-based image analysis for mapping urban land covers and quantifying buildings. It is observed that mapping heterogeneous features across large urban areas is time consuming and challenging. This study presents and demonstrates an approach for formulating an optimal land cover classification rule set over small representative training urban area image, and its subsequent transfer to the multisensor, multitemporal images. The classification results over the training area showed an overall accuracy of 96%, and the application of rule set to different sensor images of other test areas resulted in reduced accuracies of 91% for the same sensor, 90% and 86% for the different sensors temporal data. The comparison of reference and classified buildings showed ±4% detection errors. Classification through a transferred rule set reduced the classification accuracy by about 5%–10%. However, the trade-off for this accuracy drop was about a 75% reduction in processing time for performing classification in the training area. The factors influencing the classification accuracies were mainly the shadow and temporal changes in the class characteristics.     

结合边界约束网络和分水岭分割算法的建筑物提取

城市作为高密度建筑区域,在较小范围内有大量结构相似的建筑紧密分布。当前从高分辨率图像中准确检测建筑仍然是一个挑战,本文受边缘检测网络启发,提出一种强化边界精度的建筑物提取新方案,根据建筑物及边界特点改进深度网络,结合自下而上分组的分水岭分割提高分类精度和建筑边界的准确度。首先对数据预处理,生成建筑边界和建筑分割线两类辅助标签;改进性能较优的建筑检测框架ICT-Net网络,修改网络结构和损失函数,针对两类辅助标签,强化边界影响,提高网络性能;最后对网络预测结果应用结合分水岭分割和梯度提升回归树的后处理,实现高精度的建筑提取。结果表明,数据预处理、改进深度学习算法可提高建筑检测像素精度IOU (Intersection over Union)约1%。后处理能充分利用网络输出的概率信息,有效优化建筑边界,在网络预测结果的基础上提高建筑实例召回率10.5%。本文方案与原始的ICT-Net网络相比,提高建筑实例召回率22.9%。     

空地技术结合地形图测图方法研究

车载移动测量系统可以快速、高精度地对测区进行三维激光扫描,但是因地物遮挡、视角限制,使得点云数据存在缺失;无人机航测具有高效率、高灵活性和低成本等优势,但是稳定性差,受天气影像严重,易导致影像不清晰或精度低。无人机航测技术可以弥补车载移动测量技术的采集盲区,后者可以发挥高精度的优点,二者技术联合应用,将极大提高测绘精度及生产效率。本文以某小区为例,进行了相关方法实验,对建筑物顶部或植被茂密处等扫描盲区,采用无人机航测补测,通过高精度激光点云对航摄影像进行纠正匹配,综合利用激光点云与航摄影像进行大比例尺测图。     

支持向量机的建筑物激光脚点提取方法

本文提出了一种基于全色波段航空影像和激光雷达数据的建筑物检测方法.如何从激光点云数据中提取出建筑物激光脚点,是建筑物三维重建和轮廓提取的难点问题之一.植被密集区域以及与建筑物紧密相邻的树木的激光点很难与建筑物激光点区分开.本文利用支持向量机对单个激光点的特征进行两分类,特征向量包括激光点的高程、高程变化信息以及与激光点...     

面向对象的高分辨率遥感影像信息提取技术

介绍了单张高空间分辨率的航空遥感数据在没有其他专题图辅助的情况下进行面向对象的信息提取的方法。以山西省平顺县为研究区,ADS40彩色航空影像为数据源,对不同的地物采取不同尺度的多尺度分割,形成具有"同质性"的对象,然后分别采取不同的信息提取策略,得到研究区信息分类成果,且有效避免了"同谱异物"、"同物异谱"及"椒盐现象",对高分辨率遥感影像信息提取技术研究具有一定的实践指导意义。     

基于U型卷积神经网络的航空影像建筑物检测

经典的卷积神经网络结构在前向传播过程中分辨率不断下降,导致仅采用末层特征时难以实现建筑物边缘的精确分割,进而限制目标检测精度。针对该问题,提出一种基于U型卷积网络的建筑物检测方法。首先借鉴在图像分割领域中性能出色的神经网络模型U-Net的建模思想,采用对称式的网络结构融合深度网络中的高维和低维特征以恢复高保真边界;其次考虑到经典U-Net对位于特征金字塔顶层的模型参数优化程度相对不足,通过在顶层和底层两个不同尺度输出预测结果进行双重约束,进一步提升了建筑物检测精度。在覆盖范围达30 km²、建筑物目标28 000余个的航空影像数据集上的试验结果表明,本文方法的检测结果在IoU和Kappa两项关键评价指标的均值上分别达到83.7%和89.5%,优于经典U-Net模型,显著优于经典全卷积网络模型和基于人工设计特征的AdaBoost模型。     

Air photo approach to inventory of urban land use in central area case study — Kanpur

The case study of Kanpur demonstrates the use of air photo approach to collect physical data, mainly landuse in central area. The study reveals that very less percentage of vacant land exists in central area for future development. Though this area consist CBD, even then commercial landuse accounts to be only 10% of total area. Selective landuse inventory analysis was carried out for the landuse categories, commercial, industrial and vacant. Reliability of aerial photo-interpretation of over all landuse map found to be 91–95% and each sub-class of landuse reliability for omission and commission error found to be within limits for planning and design purpose.     

Monitoring the evolution of irrigated areas with Landsat images using backward and forward change detection analysis in the Kou watershed,Burkina Faso

The Kou watershed is characterized by important water resources used for drinking, agriculture (especially in the irrigated areas), industry and the preservation of aquatic fauna and flora. For several decades, there has been increasing pressure on the Kou's water resources, partly because of the expansion of the irrigated agricultural areas. This study was conducted to examine this issue, focusing on one specific irrigated area. In order to monitor the expansion of irrigated areas in developing countries, a low-cost remote sensing method based on Landsat images and aerial photographs was developed. The method is based on maximum-likelihood classification, followed by backward and forward change detection analysis requiring agronomic expertise. Using pixel trajectory analysis, the method connects all pixels to their consecutive states in order to correct their current states. The study showed that the irrigated area has expanded by almost 70% over 20 years, with most of this expansion occurring in the past 10 years. The method, if validated, could be used to obtain better information on past occupation in the rural irrigated areas for which there is currently no archived data, making temporal analyses impossible.     

加入空间-时相纹理的高分辨率图像城市建筑物物倒塌提取

将两种基于地统计学的纹理特征加入到高分辨率遥感影像的城市建筑物倒塌探测中,考察了多尺度纹理对探测结果的影响.采用基于单类支持向量机的多时相直接分类方法提取建筑物倒塌信息.以伊朗巴姆地区2003 年12 月地震前后的Quickbird 遥感影像为数据源,评价和验证了本文方法的有效性.研究表明,将多尺度的空间和时相纹理信息加入到高分辨率遥感影像的倒塌建筑物探测中,可以有效提高分类精度,该方法得到的结果可应用于灾害救援及评估.     

无人机倾斜摄影测量在房屋建筑面积测算中的应用

针对目前房屋建筑面积测量外业工作强度大、测量精度低、成图效率低等主要问题,本文提出了基于无人机倾斜摄影测量的房屋建筑面积测算方法。首先从无人机倾斜摄影、实景三维建模、数字正射影像制作、三维数字化测图、图件制作和面积指标计算等方面进行了研究,然后对试验结果进行了精度评定。结果表明,地形图平面位置中误差为4.6 cm,高程中误差为4.2 cm,房角点点位中误差为4.6 cm,建筑物边长中误差为4.5 cm,房屋建筑物面积较差小于城市商品房二级精度面积限差,各项精度均符合国家标准规范要求,该方法用于房屋建筑面积测算是可行的。     

Building population mapping with aerial imagery and GIS data

Geospatial distribution of population at a scale of individual buildings is needed for analysis of people's interaction with their local socio-economic and physical environments. High resolution aerial images are capable of capturing urban complexities and considered as a potential source for mapping urban features at this fine scale. This paper studies population mapping for individual buildings by using aerial imagery and other geographic data. Building footprints and heights are first determined from aerial images, digital terrain and surface models. City zoning maps allow the classification of the buildings as residential and non-residential. The use of additional ancillary geographic data further filters residential utility buildings out of the residential area and identifies houses and apartments. In the final step, census block population, which is publicly available from the U.S. Census, is disaggregated and mapped to individual residential buildings. This paper proposes a modified building population mapping model that takes into account the effects of different types of residential buildings. Detailed steps are described that lead to the identification of residential buildings from imagery and other GIS data layers. Estimated building populations are evaluated per census block with reference to the known census records. This paper presents and evaluates the results of building population mapping in areas of West Lafayette, Lafayette, and Wea Township, all in the state of Indiana, USA.