Moment-based feature extraction from high spatial resolution hyperspectral images

In this article, we propose a method for extracting spatio-spectral features from high spatial resolution hyperspectral (HS) images. The method is based on extracting two-dimensional moments from neighbourhoods of pixels. Three different types of moments are considered: geometric, complex Zernike and Legendre. Moments of a given type are extracted from a few principal components (PC) of HS data, and are stacked on the original HS data to form a joint spatio-spectral feature space. These features are classified using a support vector machine (SVM) classifier. The influence of the moments orders and the size of the neighbourhood window on the quality of the extracted features are analysed. A few experiments are conducted on two widely used HS data sets, Pavia University and Salinas. The results demonstrate high capabilities of the proposed method in comparison with some state-of-the-art spatio-spectral HS classification methods.     

基于XGBoost的遥感图像中道路网络的提取

传统的机器学习分类算法在处理道路提取问题过程中存在准确度低、速度慢的缺点。通过利用XGBoost算法提取遥感图像中的道路部分,以图像中的颜色特征以及像素点的局部特征作为特征输入,对图像中的每个像素点进行分类后,提取出道路。实验结果表明,该算法提取道路的准确性高,能够清晰绘出图像中的道路网络,且该算法具有较好的鲁棒性。     

基于全卷积神经网络的多源高分辨率遥感道路提取

针对半自动道路提取方法人工参与较多、提取精度不高且较为耗时的问题提出一种基于全卷积神经网络（FCN）的多源高分辨率遥感道路提取方法。首先,对高分二号和World View图像进行分割,用卷积神经网络（CNN）分类出包含道路的图像;然后,用Canny算子提取道路的边缘特征信息;最后,结合RGB、Gray和标签图放入FCN中训练,将现有的FCN模型拓展为多卫星源输入及多特征源输入的FCN模型。选取西藏日喀则地区作为研究区域,实验结果显示,所提方法在对高分辨率遥感影像进行道路提取时能够达到99.2%的提取精度,并且有效地减少了提取所需的时间。     

基于门控卷积残差网络的卫星图像道路提取

针对遥感影像中道路信息容易受到建筑物、植被等非道路信息干扰的问题,提出了一种基于门控卷积残差网络的遥感影像道路提取模型.首先,该网络使用ResNet101作为网络的编码器,在使得网络足够深的同时,也保证了梯度信息的有效传导;其次,在中心部分使用ASPP多尺度特征提取模块,进一步挖掘特征图中给予的信息;最后,使用门控卷积替换普通的卷积层,它可以根据特征图中参数的重要性,自适应分配权重,作为网络的解码器部分.该方法在CVPR DeepGlobe 2018道路提取挑战赛的数据集上进行了验证,平均交并比、Dice相似系数、召回率分别达到70.20％、82.06％、82.21％,均超过该赛事冠军DlinkNet34,提升了道路提取的效果.     

基于高分辨率SAR图像的道路自动提取

在传统算法的基础上,用多条件加权法进行道路边缘点的判断,充分利用道路的物理特性,将道路边缘点像素上下文特性作为判断的条件,以实现道路边缘线段的识别。桥接模式的思想是根据道路边缘线平行且宽度一定的特性,通过算法找出两条边缘线段之间的对应点,连接对应点以实现道路提取。经实验测试,该算法能消除地物间的影响和噪声干扰,有效地提高了道路提取的精度和速率。     

基于模糊连接度的高分辨率SAR图像道路自动提取

针对高分辨率合成孔径雷达(SAR)图像受到乘性斑点噪声的影响,且道路环境复杂多变的问题,提出一种基于模糊连接度的高分辨率SAR图像道路自动提取方法。首先,对SAR图像进行斑点滤波,以降低斑点噪声的影响;其次,结合指数加权均值比(ROEWA)算子检测结果和模糊C均值(FCM)分割结果自动提取种子点,从而提高自动化程度;最后,利用以图像灰度和ROEWA检测算子边缘强度为特征的模糊连接度算法对种子点进行扩展提取道路,经形态学处理后得到最终结果。对两幅SAR图像进行实验,并与FCM方法分割出的道路结果进行比较,所提出的方法在提取完整率、正确率及检测质量上均优于模糊C均值方法。实验结果表明,所提出的方法能较有效地从高分辨率SAR图像中提取不同宽度和弯曲程度的道路,且无需人工输入种子点。     

Urban road extraction from high-resolution optical satellite images

In recent years, many approaches have been exploited for automatic urban road extraction. Most of these approaches are based on edge and line detecting algorithms. In this paper, a new integrated system for automatic extraction of main roads in high-resolution optical satellite images is present. Firstly, a multi-scale greylevel morphological cleaning algorithm is proposed to reduce the grey deviation of the road regions. Secondly, based on the greylevel difference between road surfaces and environmental objects, a colour high-resolution satellite image is segmented into a simplified imagemap by using the mean shift algorithm, which consists of three stages. The first stage deals with image filtering, the second stage deals with colour segmentation, and the third stage is proposed to fuse small regions in the segmented image. The mean shift filter algorithm not only smoothes the image, but also preserves abrupt changes (i.e. edges) in the local structure. The mean shift segmentation algorithm is a straightforward extension of the smoothing algorithm, which preserves discontinuity. From the histogram of the simplified imagemap, we can find the potential road surfaces, and use greylevel threshold to convert the segmented image into a binary one. The binary image is processed by using binary mathematical morphological closing and opening to remove small objects and to open the connected street blocks. We use a contour tracing algorithm to remove holes in street-block regions and to detect the street blocks' contours. In this research we found that many street blocks' contours were preserved perfectly, except for some of them which were depressed. Finally, we utilize the convex hull algorithm to smooth the street blocks' zigzag edges and to close the gaps in some street blocks, and then, we get the road edges. The integrated system for road network extraction is tested on the red band of an IKONOS multispectral image; all algorithms in this study are developed in C++ under Windows XP operating system. Results of the road network extraction are presented to illustrate the validation of the extracting strategy and the corresponding algorithms in this research, and future prospects are exposed.     

A comparative study on radiometric normalization using high resolution satellite images

Remotely sensed multitemporal, multisensor data are often required in Earth observation applications. A common problem associated with the use of multisource image data is the grey value differences caused by non‐surface factors such as different illumination, atmospheric, or sensor conditions. Such differences make it difficult to compare images using the same colour metric system. Image normalization is required to reduce the radiometric influences caused by non‐surface factors and to ensure that the grey value differences between temporal images reflect actual changes on the surface of the Earth.A variety of image normalization methods, such as pseudoinvariant features (PIF), dark and bright set (DB), simple regression (SR), no‐change set determined from scattergrams (NC), and histogram matching (HM), have been published in scientific journals. These methods have been tested with either Landsat TM data, MSS data or both, and test results differ from author to author. However, whether or not existing methods could be adopted for normalizing high resolution multispectral satellite images, such as IKONOS and QuickBird, is still open for discussion because of the dramatic change in spatial resolution and the difference of available multispectral bands. In this research, the existing methods are introduced and employed to normalize the radiometric difference between IKONOS and QuickBird multispectral images taken in different years. Some improvements are introduced to the existing methods to overcome problems caused by band difference and to achieve more stable and better results. The normalized results are compared in terms of visual inspection and statistical analysis. This paper examined whether or not existing methods can be directly adopted for image normalization with high resolution satellite images, and showed how these methods can be modified for use with such images.     

Information extraction from very high resolution satellite imagery over Lukole refugee camp,Tanzania

This paper addresses information extraction from IKONOS imagery over the Lukole refugee camp in Tanzania. More specific, it describes automatic image analysis procedures for a rapid and reliable identification of refugee tents as well as their spatial extent. From the identified tents, the number of refugees can be derived and a map of the camp can be generated, which can be used for improving refugee camp management. Four information extraction methods have been tested and compared: supervised classification, unsupervised classification, multi-resolution segmentation and mathematical morphology analysis. The latter two procedures based on object-oriented classifiers perform best with a spatial accuracy above 85% and a statistical accuracy above 97%. These methods could be used for refugee camp information extraction in other geographical settings and on imagery with different spatial and spectral resolutions.     

彩色公路交通地图图像道路提取

根据彩色公路交通地图的图像特征,提出一种新颖的道路识别与提取方法。这种方法包括三个关键步骤。首先,根据区域的特征,提取出区域的灰度值;其次,根据道路的颜色和形状特征以及数字图像处理的一些方法（如对象的连通成分等）,识别并提取出道路的颜色;最后,为了获得完整的道路网络,一些道路连接方法被提出。这种算法已经被应用于许多彩色公路地图图像中去提取道路网络。大量成功的实例表明这个算法是非常有效的。     

高分辨率合成孔径雷达图像的直线特征多尺度提取方法

针对传统的合成孔径雷达（SAR）多尺度边缘提取方法中直线提取连续性和完整性不好的特点,提出了一个由粗到精的多分辨率SAR图像直线特征多级提取框架,利用多尺度策略在降低SAR图像噪声影响的同时增强相邻共线点之间的连续性和完整性,并在不同的尺度图像中根据边缘特征的特点选择不同处理方法,来实现低分辨率条件下完整直线特征的粗略提取和高分辨率的精确定位。最后用高分辨率SAR图像跑道检测实验进行了验证,并将实验结果与相位编组法和Hough变换法进行了比较。     

高分辨率遥感影像中道路震害信息的识别方法

大地震之后,紧急救援物资运输迫切需要了解灾区道路的震害信息,然而当前对遥感影像中道路的震害信息提取大多是基于像素的,提取的精度普遍不高。提出了一种面向对象的道路震害信息提取方法,通过综合利用道路的多种影像特征及震前GIS矢量道路相结合来提取道路,然后依据提取道路的完整程度来识别道路震害信息。采用汶川灾区的遥感影像为例进行了实验,与目视判读的结果比较后证明该方法有效改善了信息提取的速度和精度。     

Automated extraction of road network from medium-and high-resolution images

This paper presents an automatic methodology for road network extraction from medium-and high-resolution aerial images. It is based on two steps. In the first step, the road seeds (i.e., road segments) are extracted using a set of four road objects and another set of connection rules among road objects. Each road object is a local representation of an approximately straight road fragment and its construction is based on a combination of polygons describing all relevant image edges, according to some rules embodying road knowledge. Each road seed is composed by a sequence of connected road objects in which each sequence of this type can be geometrically structured as a chain of contiguous quadrilaterals. In the second step, two strategies for road completion are applied in order to generate the complete road network. The first strategy is based on two basic perceptual grouping rules, i.e., proximity and collinearity rules, which allow the sequential reconstruction of gaps between every pair of disconnected road segments. This strategy does not allow the reconstruction of road crossings, but it allows the extraction of road centerlines from the contiguous quadrilaterals representing connected road segments. The second strategy for road completion aims at reconstructing road crossings. Firstly, the road centerlines are used to find reference points for road crossings, which are their approximate positions. Then these points are used to extract polygons representing the contours of road crossings. This paper presents the proposed methodology and experimental results. The text was submitted by the authors in English. Aluir Porfirio Dal Poz. Year of birth: 1960. Year of graduation/Name of the institution: 1987 (Cartographic Engineering)/Sao Paulo State University. Year in which an academic degree was awarded: M.Sc. degree in Geodetic Science at Parana Federal University: 1991. Ph.D. degree in Engineering at Sao Paulo University: 1996. Affiliation: Sao Paulo State University. Position: Associate Professor. Area of research: Digital Photogrammetry and Image Analysis. Number of publications: 5 Book Chapters, 25 in Journals, and 75 in Proceedings. Membership to academies: Scientific societies: Brazilian Society of Cartography, Brazilian Society of Applied and Computational Mathematics, and Canadian Institute of Geomatics. Editorial boards and journals: Associate Editor of the Series in Geodetic Science and member of the editorial board of Brazilian Journal of Cartography. Awards and prizes for achievements in research or applications: Scientific Beginner in Cartography (1995) and Cartographic Merit (1999), both awarded by Brazilian Society of Cartography. Rodrigo Bruno Zanin. Year of birth: 1976. Year of graduation/Name of the institution: 2000 (Mathematics)/Sao Paulo State University. Year in which an academic degree was awarded: M.Sc. degree in Cartographic Sciences at Sao Paulo State University: 2004. Affiliation: Sao Paulo State University. Position: Ph.D. Candidate. Area of research: Digital Photogrammetry and Image Analysis. Number of publications: 2 in Journals and 5 in Proceedings. Giovane Maia do Vale. Year of birth: 1969. Year of graduation/Name of the institution: 1998 (Mathematics)/Sao Paulo State University. Year in which an academic degree was awarded: M.Sc. degree in Cartographic Sciences at Sao Paulo State University: 2003. Affiliation: Sao Paulo State University. Position: Ph.D. Candidate. Area of research: Digital Photogrammetry and Image Analysis. Number of publications: 3 in Journals and 10 in Proceedings. Membership to academies: Scientific societies: Brazilian Society of Applied and Computational Mathematics.     

Building change detection after earthquake using multi-criteria decision analysis based on extracted information from high spatial resolution satellite images

Change detection of ground surface objects can provide essential and precious information for experts in the fields of Geomatics, emergency management, urban management, agriculture, and forestry. Space-borne remote-sensing images are one of the main sources for change detection. Various change detection methods have been proposed on remote-sensing applications. However, often, no single efficient method can be selected for a case study because the existing methods sometimes have good performance and sometimes perform poorly. Therefore, it is necessary to propose an integrated change detection method according to some change detection methods. Multi-criteria decision analysis is a powerful framework that can integrate several criteria that may be in contrast to each other. In this study, a multi-criteria decision analysis framework was used to integrate the spectral, textural, and transformed features for detecting building changes with the help of high spatial resolution satellite images. First, the spectral, textural, and transformed features were extracted from the pre- and post-event satellite images. Second, the spectral, textural, and transformed factor maps were produced by entering the related features to three separate Adaptive Network-Based Fuzzy Inference Systems (ANFIS). Third, the ANFIS model was used again to integrate the mentioned factor maps for producing the preliminary building change map. And finally, a comprehensive sensitivity analysis was carried out to determine the proper parameters of the ANFIS models leading to accurate change detection results. The proposed method was tested on the earthquake data set of Bam City in Iran. The achieved results indicated an overall accuracy of 89.62% for identifying the changed and unchanged building regions. Moreover, the obtained results proved the efficiency and accuracy of the proposed method with respect to other implemented methods regarding the Bam earthquake. Furthermore, the aggregation of the spectral, transformed, and textural features resulted in improving the change detection accuracy by about 5–15%, compared with the accuracy of every one of them for the mentioned purpose.     

基于改进区域生长的遥感影像道路提取

提出一种改进区域生长法的遥感影像中道路提取方法。对遥感影像进行[K]均值聚类,实现道路区域和非道路区域的初步分离,并获取区域生长的基准值,按照图像特征计算出区域生长的阈值。依据对道路特性的分析,设计了9个道路路口模型。根据设计的道路路口模型,对区域生长法进行了改进,使得道路的提取按照道路路口模型自动增长。最后通过数学形态学的手段对道路进行优化。实验结果表明使用提出方法所提取道路区域更加完整。     

结合区域生长和水平集的遥感影像道路提取

提出了一种基于主成分分析(PCA)的彩色区域生长算法,并将该方法与水平集方法相结合用于高分辨率遥感影像中城市道路的提取。首先利用区域生长方法分割出大致的道路区域;然后利用预分割的结果构造初始水平集函数,进一步利用一种消除重新初始化操作的水平集方法进行道路边缘演化;最后,提出了一种不用反复初始化的水平集局部边缘修正算法,并利用该方法对因障碍物影响而错分的局部道路边界进行修正。实验结果表明,该方法能完整、有效地提取高分辨率遥感影像中的道路目标,且人工干预较少,具有较强的实用性和抗噪能力。     

Texture extraction for object-oriented classification of high spatial resolution remotely sensed images using a semivariogram

A Semivariogram, as defined in geostatistics, is a powerful tool for texture extraction of remotely sensed images. However, the traditional texture features extracted by a semivariogram are generally for pixel-based classification. Moreover, most studies have been based on the original computation mode of semivariogram and discrete semivariance values. This article describes a set of semivariogram texture features (STFs) based on the mean square root pair difference (SRPD) to improve the accuracy of object-oriented classification (OOC) in QuickBird images. The adaptive parameters for the calculation of a semivariogram were first derived from semivariance analysis, including directions, moving window size, and lag distance. Then, 22 STFs were extracted from the discrete and mean/standard deviation semivariance, and 15 features were selected from the extracted STFs based on feature optimization. Then five grey-level co-occurrence matrix (GLCM) texture features (mean, homogeneity, contrast, angular second moment, and entropy) were calculated based on segmented image objects using the panchromatic band. A comparison of classification results demonstrates that the STFs described in this article are useful supplement information for the spectral OOC, and the spectral + STFs classification method can be used to obtain a higher classification accuracy than can the combination of spectral and GLCM features.     

Auto-registration of medium and high spatial resolution satellite images by integrating improved SIFT and spatial consistency constraints

Feature-based methods have been developed in the past decades for the registration of optical satellite images. However, it is still a challenging problem to handle well the registration between medium and high spatial resolution images due to the large difference of the spatial structural features and local details for the same objects. In this study, an automated co-registration technique is proposed that integrates an improved SIFT (I-SIFT) and a novel matching strategy called spatial consistency constraints (SCC) to cope with the large difference in spatial resolutions between the image pair. Three constraints on angle, distance, and ratio are introduced to re the initial matching features obtained by I-SIFT. Three groups of experiments were conducted to validate the effectiveness of the proposed method. The experiments used high resolution multispectral and panoramic SPOT 5/6 images and Landsat 5/8 orthorectification images. Experimental results show that the registration error lies in about 1 pixel of high-resolution images and demonstrate that the proposed I-SIFT-SCC approach is suitable for fine registration of optical satellite images from medium spatial resolution to high spatial resolution with resolution ratio up to 6.     

Shadow information recovery in urban areas from very high resolution satellite imagery

It is usually quite difficult to extract and recover shadow information in the urban environment from remote sensing imagery. This paper describes the study of precisely detected shadow in satellite images and recovering information from the surface covered in shadow from very high resolution (VHR) satellite imagery.     

高分辨率遥感卫星影像的河流提取方法研究

为了克服单纯采用光谱信息提取河流的缺陷,利用高分辨率遥感影像突出的高分辨率的特性提出一种综合影像中光谱、纹理、几何特性等多特征联合提取河流的方法。该方法分别对河流水体的光谱特征、纹理特征及河流几何形状进行描述,选取特征参数,构造综合特征矩阵,利用均值聚类分割最终得到河流目标。通过对真实高分辨率遥感影像Worldview1影像进行的实验验证了该方法的高精准性及快速性。