基于纹理-模糊连接度的遥感影像道路自动提取

提出一种基于纹理-模糊连接度的遥感影像道路信息全自动提取算法。利用Canny算子对影像进行边缘检测,通过道路阈值和分类算法的联合优化,提取出道路种子点。定义了包含纹理能量的模糊连接度,对影像中各像素相对于多种子点的连接度进行遍历计算,寻址最优路径,从而识别出完整的道路。实验结果表明,该道路提取算法相对于模糊连接度算法精度更高,具有较强的通用性。     

基于变差函数纹理和BP人工神经网络的QuickBird影像分类研究

通过对QuickBird影像的主成分分析,发现第一主成分信息量为67.09%,第二主成分信息量为31.73%,合计98.81%,因此选了第一、第二主成分进行纹理提取。纹理提取时采用绝对变差函数纹理,窗口大小为3×3,步长为1。地物类型分为4类即建筑物、水体、植被和裸地。BP人工神经网络的拓扑结构为4-4-1,隐层传递函数为S函数(logsig),输出层传递函数为线性(purelin),训练函数为Trainscg;应用该神经网络结构分别对QuickBird影像的多光谱影像和一二主成分及一二主成分纹理影像进行了分类,分别算出了前者的P=84.54%,Kappa系数K=78.50%;后者的P=89.46%,Kappa系数K=85.29%,同时发现加入纹理后分类结果显著提高。     

结合非监督分类和几何—纹理—光谱特征的高分影像道路提取

目的 目前针对复杂场景高分辨率遥感影像道路提取多采用监督分类方法,但需要人工选择样本,自动化程度低且具有不稳定性。基于像元级的方法,提取完整度低且易产生椒盐噪声;面向对象的方法易产生粘连问题。为了提高道路提取的完整度、准确度和自动化程度,提出一种基于非监督分类和几何—纹理—光谱特征的道路提取方法。方法 首先考虑光谱特征利用非监督分类进行初步分割,结合基于纹理特征分类的结果得到初始道路区域。然后根据道路特征建立一套完整的非道路区域滤除体系：边缘滤波断开道路和非道路的连接、纹理滤波滤除大面积非道路区域、形状滤波去除剩余小面积非道路区域。最后利用张量投票算法得到连贯、平滑的道路中心线。结果 选择复杂场景下的高分辨率IKONOS影像和QuickBird影像进行实验,与国内外基于像素和面向对象的两种有代表性的道路提取方法进行对比,采用完整率、正确率、检测质量3个评价指标进行定量评价。实验结果表明该方法相比于其他算法在完整率、正确率和检测质量上平均提高26.61%、5.57%和26.77%。定性分析结果表明,本文方法可以有效改善椒盐噪声和粘连现象。此外本文方法的自动化程度更高。结论 提出了一种基于非监督分类和几何—纹理—光谱特征的高分辨遥感影像道路提取方法,非监督相对于监督分类的方法有更高的自动化程度,复杂场景下的道路提取融合几何—纹理—光谱特征有效避免了基于像元级道路提取易产生的椒盐噪声现象和面向对象道路提取易产生的粘连现象。该方法适用于高分辨率遥感影像城市道路提取,能够得到较高的完整度、准确度以及自动化程度。非监督分类和多特征结合的道路提取方法有广阔的应用前景。     

可见光红外成像辐射仪数据林火识别算法研究

针对新型传感器可见光红外成像辐射仪(VIIRS)数据具有较宽的波段覆盖范围、中等空间分辨率以及高重访周期,可以在火灾监测中发挥重要作用的特点,该文结合VIIRS数据的波段设置,分析了火点像元与背景像元在反射波段的地表反射率差异及热红外波段的亮度温度差异,研究了应用该数据进行林火识别的算法。并对2013年8月发生在美国加州斯坦尼斯劳斯国家森林公园的火灾,开展了基于该算法的应用研究,使用中高分辨率Landsat ETM+/OLI影像对提取结果进行了验证。结果表明,该文提出的林火识别模型在火点集中区域可达到85%的精度,较好地提取了火灾发生的区域分布。     

利用Deeplab v3提取高分辨率遥感影像道路

针对传统道路提取方法存在的道路边缘粗糙、抗干扰性弱、提取精度低等问题,提出了一种基于编码解码器的空洞卷积模型(Deeplab v3)的道路提取方法.首先,对原始高分辨率遥感影像进行标注;其次,利用标注数据集对Deeplab v3模型进行训练、测试;最后,得到高分辨率遥感影像道路提取结果.分析结果可知,该模型能够较好地提...     

分割道路影像并形成道路网的新方法

根据卫星影像提取道路信息过程,提出了一种从高分辨率卫星影像中提取道路网的新方法.先找到一点处纹理和灰度一致性最优方向,以此方向上的纹理描述值构成分割用特征矢量,将MRF模型应用于特征空间分割出道路目标.接着用改进的直线段匹配法去除以错分点组成的斑块,并借助仿真实验手段取得道路段中心线.连接中心线形成道路网.仿真结果验证了新方法的正确性.     

基于Swin Transformer的遥感影像道路提取方法

在先进的交通系统中,道路提取是最重要的任务之一。高分辨率遥感影像道路区域的提取具有复杂的背景和道路网络的异质性、高类间差异和低类内差异等特点。近几年来,卷积神经网络(CNN)在道路提取方面取得了里程碑式的进展。虽然CNN已经取得了很好的发展,但是由于卷积运算的局域性,网络无法很好地学习全局和长程语义信息交互。本文提出了Swin Transformer Unet,它结合了带有跳跃连接的U型编解码器结构和带有移位窗口的Swin Transformer模块。为了获得更好的性能,本文采用了数据增广、数据预处理等技术。本文选取马萨诸塞州道路数据集作为数据集进行道路提取实验,结果表明,所提出的网络在遥感图像道路提取中的性能优于其他U形网络,可以实现遥感影像道路的精确提取。     

基于波段运算和纹理特征的高分一号多光谱数据云检测

针对国产高分一号卫星16m宽幅多光谱遥感影像仅含有可见近红外波段的情况,提出了基于波段运算与空间纹理特征相结合的云检测方法。首先基于光谱特征和波段运算生成初始云掩膜,然后结合纹理特征提高检测精度,最后采用阈值分割生成云掩膜。实验结果表明,该方法能够很好地检测出不同时相、不同类型下垫面上空的云像元,可以满足应用需要。由于在云检测方法中综合应用了波段光谱和空间纹理特征,理论上具有一定的普适性,对其他仅有可见近红外波段的遥感影像云检测具有参考意义。     

山区可见光—近红外遥感影像浓密植被暗像元自动识别方法研究

暗目标法是目前气溶胶光学厚度遥感反演中应用最为广泛的方法,浓密植被暗像元的识别是暗目标法的基础。针对可见光—近红外影像缺少中红外波段难以有效识别浓密植被暗像元的问题,引入红波段直方图阈值法识别山区可见光—近红外影像的浓密植被暗像元。该方法利用浓密森林像元在可见光波段反射率低的特点,通过搜索红波段直方图的最小峰值自动识别浓密植被暗像元。试验中选取Landsat TM影像前4个波段利用红波段直方图阈值法识别可见光—近红外影像的浓密植被暗像元,并与在中红外波段影像和可见光—近红外影像中广泛应用的两种暗像元识别方法进行对比分析,探讨红波段直方图阈值法的有效性,最后将该方法应用于环境减灾卫星(HJ-1)CCD影像的暗像元识别和气溶胶反演。实验结果表明:红波段直方图阈值法明显优于常用的可见光—近红外影像暗像元识别方法,识别精度接近传统的中红外波段影像识别方法,相似度指数小于2和小于3的暗像元分别为83.12%和93.48%。该方法为山区可见光—近红外影像浓密植被暗像元自动识别提供了一种新的适用方法,识别结果能够满足暗目标法反演气溶胶光学厚度的要求。     

利用邻域质心投票从分类后影像提取道路中心线

目的 利用分类算法对高分辨率影像中的道路进行分割时,得到的二值图像往往混杂了许多非道路区域,且道路区域呈面状,无法直接应用于生产与研究。针对该问题,提出一种利用邻域质心投票提取道路中心线的算法。方法 首先检测像素在各方向上的连通距离以构建邻域多边形,随后进行质心投票来提取道路的中心线,与此同时估算道路宽度并判断出连通距离较长的方向数目,以排除非道路区域的干扰,最后经形态学处理得到细化的中心线。结果 选取测试图像及具有不同道路分布特征的高分辨率航空影像的分类结果进行实验,并将该算法与Zhang和Couloigner提出的算法进行了对比分析。结果显示,该算法的提取质量为80.6%和79.0%,且计算效率较高,处理实际影像的用时小于参考算法的20%,此外在稳定性及对不同路宽的适应性等多个方面均具有优势。结论 提出一种邻域质心投票算法,该算法能够同时实现传统方法中提纯与中心线提取两个步骤所对应的功能,从分类影像直接提取道路中心线。实验结果表明,该算法能够根据形状特征有效检测道路,且具备一定抗干扰能力,适用于对混杂了非道路区域的高分辨率影像的分类结果进行处理。     

从高分辨率遥感影像中提取城市道路的新方法

在综合几种现有算法优点的基础上,提出一种新的道路提取策略。首先以角度纹理特性法分割原始影像;接着利用直线匹配原理剔除初始分割结果中的非道路地物,得到更为规则的道路条带;然后通过形态学手段获得道路中心线,并将每条中心线拆分为多段直线;结合上下文知识的马尔可夫模型被用于组织道路段的中心线,从而恢复完整道路网。实验结果表明：新方法具有良好的性能,可以从高分辨IKONOS遥感影像中提取出复杂的城市道路。     

结合像元形状特征分割的高分辨率影像面向对象分类

针对高分辨率遥感影像空间分辨率高,结构形状、纹理、细节信息丰富等特点,提出一种新的融合特征的面向对象影像分类方法来提取城市空间信息。基本过程包含以下4个方面:①提取影像的几何纹理等结构;②融合几何与纹理特征的面向对象影像分割;③提取对象的形状、纹理和光谱特征,并优选最佳特征子集;④最后基于支持向量机(SVM)完成面向对象的影像分类。通过对福州IKONOS影像数据实验,结果表明融入影像特征后的分割效果明显优于原始影像的分割结果,而信息最大化(mRMR)的特征选择能够快速地获得较好的特征子集。通过与eCognition最邻近分类方法比较,表明本文方法的分类总体精度大约提高了6%,效果显著。     

Mapping health levels of Robinia pseudoacacia forests in the Yellow River delta,China, using IKONOS and Landsat 8 OLI imagery

The largest artificial Robinia pseudoacacia forests in the Yellow River delta of China have been infected by dieback diseases. Over the past several decades, this has caused a large amount of mortality of Robinia pseudoacacia forests in this area. Timely and accurate information on the health levels of the forests is crucial to improving local ecological and economic conditions. Remote sensing has been demonstrated to be a useful tool to map forest diseases over a large area. In this study, IKONOS and Landsat 8 Operational Land Imager (OLI) sensor data were collected for comparing their capability of accurately mapping health levels of the artificial forests. There were three health levels (i.e. healthy, medium dieback, and severe dieback) based on explicit tree crown symptoms. After the IKONOS and OLI images were preprocessed, both spatial and spectral features were extracted from the IKONOS and OLI imagery, and a maximum likelihood classification method was used to identify and map health levels of Robinia pseudoacacia forests. The experimental results indicate that the IKONOS sensor has greater potential for identifying and mapping forest health levels. Furthermore, texture features, especially texture variance, derived from the IKONOS panchromatic band, contributed greatly to the accuracy of classification results, achieving an overall accuracy (OA) of 96% for the IKONOS sensor and an OA of 88% for the OLI 2, which used both OLI spectral and IKONOS spatial features, compared with an OA of 74% for the OLI sensor alone. Our results indicate that the texture features extracted from high resolution imagery can improve the classification accuracy of health levels of planted forests with a regular spatial pattern. Our experimental results also demonstrate that classification of an image with a spatial resolution similar to, or finer than, tree crown diameter outperforms that of relatively coarse resolution imagery for differentiating living tree crowns and understorey dense green grass.     

Sub-pixel estimation of urban land cover components with linear mixture model analysis and Landsat Thematic Mapper imagery

We examine the utility of linear mixture modelling in the sub-pixel analysis of Landsat Enhanced Thematic Mapper (ETM) imagery to estimate the three key land cover components in an urban/suburban setting: impervious surface, managed/unmanaged lawn and tree cover. The relative effectiveness of two different endmember sets was also compared. The interior endmember set consisted of the median pixel value of the training pixels of each land cover and the exterior endmember set was the extreme pixel value. As a means of accuracy assessment, the resulting land cover estimates were compared with independent estimates obtained from the visual interpretation of digital orthophotography and classified IKONOS imagery. Impervious surface estimates from the Landsat ETM showed a high degree of similarity (RMS error (RMSE) within approximately ±10 to 15%) to that obtained using high spatial resolution digital orthophotography and IKONOS imagery. The partition of the vegetation component into tree vs grass cover was more problematic due to the greater spectral similarity between these land cover types with RMSE of approximately ±12 to 22%. The interior endmember set appeared to provide better differentiation between grass and urban tree cover than the exterior endmember set. The ability to separate the grass vs tree components in urban vegetation is of major importance to the study of the urban/suburban ecosystems as well as watershed assessment.     

Previously trained neural networks as ensemble members: Knowledge extraction and transfer

The use of Artificial Neural Networks (ANNs) for the classification of remotely sensed imagery offers several advantages over more conventional methods. Yet their training still requires a set of pixels with known land cover. To increase ANN classification accuracy when few training data are available, an algorithm was applied that allows experience gained in previous classifications to be reused. The proposed method was evaluated by classifying a tropical savannah region in northern Togo using Landsat Thematic Mapper (TM) imagery. The presented approach reached a mean kappa coefficient that was significantly larger (at the 95% level) than that obtained after training networks with randomly initialized weights. Also, the observed variances on the obtained accuracies were significantly lower when compared to networks that were randomly initialized. Finally, Bhattacharyya (BH) distances were used to explain why some land cover classes benefit more from knowledge transfer than others.     

基于遥感光谱和空间变量随机森林的黄河三角洲刺槐林健康等级分类

对刺槐林健康状况进行准确分类制图,是进行刺槐林健康状况评估与生态修复的前提。以高分辨率IKONOS影像、基于影像提取的不同窗口、不同灰度共生矩阵纹理信息以及反映局部空间自相关的Local Getis-Ord Gi(Getis统计量)为数据源,结合实测生态样方数据,利用多决策树的组合分类模型随机森林(RF)对刺槐林健康进行分级,对6种方法的分类精度进行了比较且对分类变量的重要性进行了排序。结果显示:19m×19m是最佳纹理计算窗口;灰度共生矩阵均值是最优纹理变量;基于波段4计算的Getis统计量对RF分类具有最重要的作用;较之利用全部光谱、纹理和Getis统计量的80个波段/变量,利用前向选择得到的前16个重要性变量进行RF分类,获得了最高的分类精度(总精度为93.14%,Kappa系数为0.894)。研究证实了从高分影像提取的空间特征信息有助于提高对具有规则分布格局的人工刺槐林健康等级的分类精度;前向选择方法可以利用较少的预测变量获得较高的分类精度。     

Quantifying high‐resolution impervious surfaces using spectral mixture analysis

Impervious surface distribution and its temporal changes are considered key urbanization indicators and are utilized for analysing urban growth and influences of urbanization on natural environments. Recently, urban impervious surface information was extracted from medium/coarse resolution remote sensing imagery (e.g. Landsat ETM+ and AVHRR) through spectral analytical methods (e.g. spectral mixture analysis (SMA), regression tree, etc.). Few studies, however, have attempted to generate impervious surface information from high resolution remotely sensed imagery (e.g. IKONOS and Quickbird). High resolution images provide detailed information about urban features and are, therefore, more valuable for urban analysis. The improved spatial resolution, however, also brings new challenges when existing spectral analytical methods are applied. In particular, a higher spatial resolution leads to reduced boundary effects and increased within‐class variability. Taking Grafton, Wisconsin, USA as a study site, this paper analyses the spectral characteristics of IKONOS imagery and explores the applicability of SMA for impervious surface estimation. Results suggest that with improved spatial resolution, IKONOS imagery contains 40–50% of mixed urban pixels for the study area, and the within‐class variability is a severe problem for spectral analysis. To address this problem, this paper proposes two approaches, interior end‐member set selection and spectral normalization, for SMA. Analysis of results indicates that these approaches can reasonably reduce the problems associated with boundary effects and within‐class variability, therefore generating better impervious surface estimates.     

A contextual classification scheme based on MRF model with improved parameter estimation and multiscale fuzzy line process

A Markov random field (MRF) based method using both contextual information and multiscale fuzzy line process for classifying remotely sensed imagery is detailed in this paper. The study area known as Elkhorn Slough is an important natural reserve park located in the central California coast, USA. Satellite imagery such as IKONOS panchromatic and multispectral data provides a convenient way for supporting the monitoring process around this area. Within the proposed classification mechanism, the panchromatic image, benefited from its high resolution, mainly serves for extracting multiscale line features by means of wavelet transform techniques. The resulting multiscale line features are merged through a fuzzy fusion process and then incorporated into the MRF model accompanied with multispectral imagery to perform contextual classification so as to restrict the over-smooth classification patterns and reduce the bias commonly contributed by those boundary pixels. The MRF model parameter is estimated based on the probability histogram analysis to those boundary pixels, and the algorithm called maximum a posterior margin (MPM) is applied to search the solution. The results show that the proposed method, based on the MRF model with the multiscale fuzzy line process, successfully generates the patch-wise classification patterns, and simultaneously improved the accuracy and visual interpretation.     

Remote detection of invasive Melaleuca trees (Melaleuca quinquenervia) in South Florida with multispectral IKONOS imagery

The distribution of invasive Melaleuca (Melaleuca quinquenervia (Cav.) S.T. Blake) was mapped using 4‐m spatial resolution, multispectral IKONOS imagery in an area of south Florida along the eastern edge of Everglades National Park. Detection of Melaleuca stands was achieved using a back‐propagation neural network classifier, which allowed identification of dense stands, but in some instances misclassified other woody canopies as Melaleuca. The use of IKONOS multispectral imagery to detect low‐density occurrences of Melaleuca appears limited relative to traditional methods of aerial photographic interpretation. However, analysis of landscape‐level distribution of moderate‐to‐dense Melaleuca, using Fragstats, indicated a highly aggregated Melaleuca distribution relative to other woody vegetation patches. The distribution of Melaleuca stands was associated with cultural features in the suburban environment such as canals and roads, which may act as dispersal corridors for seeds. Thus, classified IKONOS imagery may be useful for inferring landscape patterns that relate to the persistence and spread of Melaleuca and other invasive species.     

Spatial context-dependent multi-scale and directional image texture

This study evaluated image texture within a multi-scale context. The multi-scale texture profile is a representation of texture as a function of scale. It is related to the variogram, but has the advantage of being a direct measure of local texture over a specified kernel. Directional texture is a method to capture the dependence of texture on orientation using 1-dimensional kernels, and is potentially useful for quantifying texture for classes with a linear shape, such as roads or rivers. Seven derived texture attributes were proposed from the multi-scale texture profile and directional texture: the minimum, median, mean, maximum, range, minimum directional, and minimum multi-scale and directional texture. The derived texture attributes were found to be useful in developing a locally adaptive texture measure, which uses different texture attributes for different locations in the image. In order to develop the rules for the adaptive method, seven characteristic multi-scale texture profiles were identified. Rules were established to identify the general classes, and each pixel was assigned to one class only, depending on that pixel's texture profile. Optimal texture measures were proposed for each class. In an analysis using selected texture classes from IKONOS test data from Jeju, Korea, it was found that compared to all other texture measures, the adaptive texture provided the highest average and minimum classification accuracy, as well as the second highest maximum accuracy (69%, 51% and 92%, respectively). In comparison, the best fixed size kernel, 11?×?11 pixels, had accuracies in the same categories of 68%, 38% and 87%, and the original panchromatic image band had accuracies of 41%, 5% and 92%. Future work is needed to enhance the texture rules, and apply the adaptive texture method to other scales of data.