基于分水岭变换的遥感影像面向对象多尺度分割算法研究

面向对象多尺度分割技术建立一种基于多尺度分割、逐级分层提取地物边界的方法。并基于这一思想,提出基于分水岭算法和异质性最小区域合并算法相结合的快速分割方法。试验表明,该方法能快速准确地获得高分辨率遥感图像的分割结果。     

基于eCognition的遥感图像面向对象分类方法研究

随着高分辨率遥感图像越来越普及,传统的面向像元的图像分类方法不能满足对高分辨率遥感图像区域分类的需求,高分辨率遥感图像对图像处理的软件与硬件都有了更高的要求,因此,出现了相较于面向像元有着更高精度更为合理的面向对象分类方法,也更加适用于高分辨率遥感影像。本文通过采用面向对象分类的基本方法,运用eCognition软件,以山东省胶州市地区遥感影像为例,进行多尺度分割和面向对象分类。并用ENVI做监督分类,基于目视解译精度评定,对不同方法作出分析评价。结果表明:面向对象分类方法精度更高,更具有可靠性。     

高分一号多光谱遥感数据的面向对象分类

文章针对高分一号(GF-1)高分辨率遥感数据,提出了一种基于多特征的面向对象遥感图像分类算法:首先,对GF-1卫星数据进行分水岭分割,并利用仿射不变矩形状特征算子获得遥感图像的几何特征;其次,利用主成分分析和灰度共生矩阵获得遥感图像的纹理特征;然后,基于多特征数据进行均值漂移滤波,并利用自动标记分水岭分割方法实现遥感图像分割;最后,结合基于像元的最大似然监督分类结果做投票分类处理,从而实现面向像元与面向对象相结合的遥感数据分类.以高分一号遥感数据进行分类实验,结果表明:本文方法可有效地提高遥感图像分类精度.     

基于相位一致的高分辨率遥感图像分割方法

基于分水岭变换的图像分割性能在很大程度上依赖于用来计算待分割图像梯度的算法。根据频域相位信息对图像特征的表征能力,引入相位一致的思想计算图像特征,应用Log Gabor小波提取高分辨率遥感图像的多尺度梯度。接着在对相位一致梯度进行分水岭分割时发现,在抑制分水岭算法的过度分割方面,经典的基于前景标记和背景标记的方法并不适合于遥感图像的分割,给出一种基于前景标记和梯度重建的分水岭算法。对IKONOS Pan图像上的农田、厂房和居民楼等地物进行特征提取和图像分割实验,结果表明相位一致方法优于空域特征检测算子,根据相位一致特征得到较好的分水岭分割结果。     

结合异质性及分水岭变换的遥感影像分割方法

面向对象的遥感信息提取,首要任务是对遥感影像进行分割,分割的目的在于把原始图像分割成一些在空间上相邻、光谱相似的同质区域。传统的分水岭变换对噪声敏感,易产生过分割现象,不能直接适用于遥感影像的信息提取。本文利用非线性同组滤波方法(PGF)消除原始影像噪声,根据高分辨率遥感影像地物的异质性特征,对分水岭初始分割结果进行异质性准则合并。实验结果表明,该方法对高分辨率遥感影像的分割效果良好,能够适用于面向对象的信息提取。     

面向对象的成都平原多源遥感影像分割尺度研究

要对高分辨率遥感影像进行分类,采用面向对象的遥感影像分析技术比传统的面向像元的遥感影像分析技术优越。要使用面向对象的遥感影像分析技术,关键的第一步是要对遥感影像进行分割,以便得到一系列与地物有密切联系的影像对象。分割的准确性与分割的尺度选择有关。本文针对成都平原高分辨率卫星影像分割尺度选择进行试验和研究,采用不同尺度对试验区不同分辨率遥感影像进行影像分割,并比较分割结果,得出成都平原高分辨率遥感影像数据分割最佳尺度与影像对象亮度均值标准差最大值所对应的分割尺度一致;并且遥感影像空间分辨率越高,最佳分割尺度越大,反之亦然。     

高分辨率遥感图像频域特征提取与图像分割研究

遥感图像分割是面向对象识别的关键,论文基于频谱分析理论与频域滤波方法,对IKONOS卫星图像的特征提取与图像分割技术进行了研究,根据振幅和相位信息提取遥感图像的纹理和边缘特征,并提出多频段标记算法实现了结合纹理和边缘特征的图像分割.论文主要研究内容与结论包括:     

成都平原高分辨率遥感影像分割尺度研究

对高分辨率遥感影像进行分类,面向对象的影像分析技术优于传统的面向像元的影像分析技术。要使用面向对象的影像分析技术,关键是分割遥感影像,以得到一系列与地物有密切联系的影像对象。分割的准确性与分割的尺度选择有关。针对成都平原高分辨率卫星影像,采用不同尺度对试验区影像进行分割,并比较分割结果,确定成都平原高分辨率遥感影像数据分割最佳尺度为30,该尺度分割的影像对象亮度均值标准差最大。     

一种基于相位一致的高分辨率遥感图像特征检测方法

精确检测图像边缘特征是进行高分辨率遥感图像分割和识别的关键。空域特征检测算子以解决阶跃形边缘为主,得到的边缘特征对图像的亮度和对比度敏感。本文引入了一种基于频域相位一致的图像特征检测方法,该方法对遥感图像亮度和对比度具有不变性,同时适用于多种边缘特征的检测。使用Log Gabor小波计算IKONOS Pan图像的相位一致多尺度梯度,对农田、道路和厂房等典型地物进行特征检测的结果表明,相位一致算法对图像局部亮度和对比度不敏感;并且对线形物体产生单线响应,不似空域检测算子产生双线响应。最后考察滤波器尺度和方向参数变化及添加高斯噪声对检测结果的影响,发现相位一致算法无需先使用低通滤波去除噪声,因而具有更稳定的特征定位精度;并且抗噪声干扰的能力强,检测结果不会因为噪声而出现波动。基于相位一致的遥感图像不变特征提取,为高分辨率遥感图像的分割和对象识别提供了基础。     

深度神经网络条件随机场高分辨率遥感图像建筑物分割

高分辨率遥感图像建筑物分割的实质是构建一个输入图像到分割结果之间的高维强非线性映射模型。然而,建筑物可能遍布整幅遥感图像,则在语义分割过程中,当前像素点可能与非邻域的像素点存在直接关系。为了更加精确地逼近建筑物分割的真实映射模型,克服道路、建筑物错层和阴影的影响,提高分割精度,本文以深度残差神经网络为基础,构建Encoder-Decoder的深度学习架构,自动提取建筑物的特征,学习建立高维强非线性分割模型;同时,通过条件随机场的成对势函数调节当前像素点与其他像素点之间的关联关系,从而构成全连接条件随机场对Encoder-Decoder的分割结果进行调节,提升分割精度。在全连接条件随机场的计算过程中,采用循环神经网络的运行机制来完成均值场的计算,这将条件随机场与深度神经网络有机融合,实现了Encoder-Decoder和全连接条件随机场参数的同步训练。实验结果表明,本文采用的深度神经网络条件随机场方法能有效克服道路、建筑物错层和阴影的影响,提升高分辨率遥感图像中建筑物的分割精度;同时,在一定范围内对多分辨率遥感图像具有较好的泛化能力。     

航空数字摄影测量传感器系统新进展

近些年来,随着数字成像技术的快速发展,大幅面框幅式数字航空成像系统和机载三线阵成像系统正得到迅速推广和应用,本文根据数字成像系统的新特点及国内外主要研究方向,参考第21届国际摄影测量与遥感大会中关于该专题的49篇文章,对航空数字成像系统的数据处理与系统集成、Leica ADS40系统、VEXCEL/Microsoft UltraCamD/UltraCamX系统、Intergraph DMC系统、无人机（UAV）数字航空成像系统等研究方向进行了总结。     

Image Sources for Informal Settlement Management

Informal settlements are a common feature of developing countries. In South Africa the improvement of living conditions in these settlements and the upgrading to formal housing types are regarded as being of central importance to the nation's development. Effective settlement improvement and upgrading activities, which we here term "management", require adequate spatial data. To date, the acquisition of spatial models of informal settlements has been based on conventional mapping techniques, and mostly on photogrammetry. Data are compiled using analogue or analytical methods. These are manual and hence require both considerable expertise and expensive equipment. Moreover, these methods are uneconomical over the often relatively small, densely populated areas covered by informal settlements and are also too expensive to employ with a regularity required to support such tasks as change detection. Alternative imaging sources and mapping techniques are therefore needed.
In this article we examine the problem of spatial information acquisition for informal settlement management from three perspectives: spatial information requirements, the role which imagery can play in satisfying these spatial information requirements, and effective imaging options. We focus on the potential of high resolution satellite imaging, small format digital aerial imagery and digital multispectral video systems for rapid mapping. We also discuss the example of automated shack extraction from aerial imagery.     

线阵列卫星传感器定向方法的研究

介绍了确定线阵列卫星影像外方位元素的几种常用方法 ,通过试验对各种定向方法所能达到的地面点测定精度作了比较     

Orthophotograph Production in Urban Areas

This paper investigates possible solutions to the problem of urban orthophotography. Extensive work has been carried out to attain the best possible results, using all the existing software in the Department of Geomatic Engineering, University College London. Two of the most promising methods, which yield the best results, are analysed further. New routines were written in C language, as additions to the existing software, to allow for the special problems that arise in urban areas. The final results overcome the problem of double mapping (or occlusions) and will potentially enable even greater success to be achieved. A simple accuracy assessment of the results has been made. Difficulties and current limitations are analysed and presented.     

高分辨率卫星遥感数据在城市地图更新中的应用

城市地图是全面反映城市发展和城市社会经济现象的综合信息载体,是城市基础空间数据库和城市GIS的重要组成部分。文章就高分辨率卫星遥感数据在城市地图更新中的工艺、生产平台、影像纠正、影像融合、地图更新的方法及应用进行了探讨。     

Multi-temporal RADARSAT-2 polarimetric SAR for maize mapping supported by segmentations from high-resolution optical image

Due to its ability to penetrate the cloud, Synthetic Aperture Radar (SAR) has been a great resource for crop mapping. Previous research has verified the applicability of SAR imagery in object-oriented crop classification, however, speckle noise limits the generation of optimal segmentation. This paper proposed an innovative SAR-based maize mapping method supported by optical image, Gaofen-1 PMS, based segmentation, named as parcel-based SAR classification assisted by optical imagery-based segmentation (os-PSC). Polarimetric decomposition was applied to extract polarimetric parameters from multi-temporal RADARSAT-2 data. One Gaofen-1 image was then used for parcel extraction, which was the basic unit for SAR image analysis. The final step was a multi-step classification for final maize mapping including: the potential maize mask extraction, pure/mixed maize parcel division and an integrated maize map production. Results showed that the overall accuracy of the os-PSC method was 89.1%, higher than those of pixel-level classification and SAR-based segmentation methods. The comparison between optical- and SAR-based segmentation demonstrated that optical-based segmentation would be better at representing maize field boundaries than the SAR-based segmentation. Moreover, the parcel- and pixel-level integrated classification will be suitable for many agricultural systems with small landownership where inter-cropping is common. Through integrating advantages of the SAR and optical data, os-PSC shows promising potentials for crop mapping.     

Rectification of Slant Range Imagery Through a Direct Image to Ground Relationship

A theoretical model for describing a direct image to ground relationship for slant range images extracted from synthetic aperture radar imaging satellites (for instance ERS-1 and ERS-2) is formulated. The rigorous orbital model is based on orbital parameters. Transformations and equations describing the relationship are derived.     

Mapping deer habitat suitability using remote sensing and geographic information systems

This paper describes an effort to map the habitat for the Eastern Tehama Deer Herd located in Northern California. The range of this herd encompasses almost 600,000 hectares (1.5 million acres). Knowledge of the spatial distribution of suitable habitat is prerequisite to managing the deer herd. Remote sensing and GIS are appropriate tools for such an assessment. Remotely sensed data were used to map vegetation/land cover. These data along with elevation, aspect, slope, juxtaposition, and various buffer zones were input into a GIS and a model was used to produce a map of habitat suitability. The accuracy of the vegetation/land cover map was assessed and methods for validating the habitat suitability map are presented.     

月球卫星三线阵CCD影像EFP光束法空中三角测量

本文对月球卫星摄影三线阵CCD影像的EFP(等效框幅相片)光束法空中三角测量作两种处理:一是与现行摄影测量常用的将平差转到切面坐标系进行,二是在摄影测量坐标系内,长航线自由网EFP光束法平差利用三线CCD推扫特点,在EFP平差中增加对前、后视影像的相机主距的附加改正项,用以补偿由于球面曲率产生的前、后视影像比例尺的差异,平差得到的是平面基准的地面坐标及外方位元素的平差值。前者计算,数学上严格,但长航线要适当分段为切面处理;后者计算数学上有近似性,可方便地用于估算卫星姿态变化率,或作地面模型的几何反演等实验研究。利用嫦娥一号获取的第一条航线,并给出相应的结果。     

ArcGIS ModelBuilder模型设计方法在影像解译中的应用

在影像解译的工作过程中,需要对大量的栅格和矢量数据进行处理,绝大部分的数据处理工作是一种流程化的重复操作。ModelBuilder是一个ArcGIS平台下一个用来创建、编辑和管理模型的应用程序。文中利用ModelBuilder快速地构建模型,模型根据设定的程序自动处理数据,在一定程度上提高影像解译工作的效率,保证最终成果质量。