Improved maize cultivated area estimation over a large scale combining MODIS–EVI time series data and crop phenological information

The accurate and timely information of crop area is vital for crop production and food security. In this study, the Enhanced Vegetation Index (EVI) data from MODerate resolution Imaging Spectroradiometer (MODIS) integrated crop phenological information was used to estimate the maize cultivated area over a large scale in Northeast China. The fine spatial resolution China’s Environment Satellite (HJ-1 satellite) images and the support vector machine (SVM) algorithm were employed to discriminate distribution of maize in the reference area. The mean MODIS–EVI time series curve of maize was extracted in the reference area by using multiple periods MODIS–EVI data. By analysing the temporal shift of crop calendars from northern to southern parts in Northeast China, the lag value was derived from phenological data of twenty-one agro-meteorological stations; here integrating with the mean MODIS–EVI time series image of maize, a standard MODIS–EVI time series image of maize was obtained in the whole study area. By calculating mean absolute distances (MAD) map between standard MODIS–EVI image and mean MODIS–EVI time series images, and setting appropriate thresholds in three provinces, the maize cultivated area was extracted in Northeast China. The results showed that the overall classification accuracy of maize cultivated area was approximately 79%. At the county level, the MODIS-derived maize cultivated area and statistical data were well correlated (R² = 0.82, RMSE = 283.98) over whole Northeast China. It demonstrated that MODIS–EVI time series data integrated with crop phenological information can be used to improve the extraction accuracy of crop cultivated area over a large scale.     

高分三号影像水体信息提取

国内外针对陆地水体信息提取、洪涝灾害快速响应方面具有较深入的研究,但是多采用发展较早、图像质量可靠的可见光影像及国外星载SAR影像。中国合成孔径雷达（SAR）卫星高分三号（GF-3）已获取了大量多极化、全极化SAR数据,为了将GF-3影像快速应用到环境保护、水资源管理等行业中,本研究分析了水体与其他目标具有的不同后向散射特性,将阈值分割法与马尔可夫随机场（MRF）相结合,发展了一种检测精度较高、自动化程度强的水体信息提取方法。该方法首先通过直方图统计的方法对不同成像模式、不同极化的GF-3影像进行后向散射强度分析,在阈值分割的研究基础上,比较了最大类间方差法（Otsu）和Kittler and Illingworth（KI）二值化法在水体-非水体分类中的效果。然后结合DEM和GF-3轨道参数排除因阴影现象产生的辐射失真对图像概率分布的影响,得到初始的水体信息分布图,再经过Fisher变换和马尔可夫随机场（MRF）的迭代运算,综合利用GF-3影像的多极化信息和空间上下文信息,以最大后验概率准则输出最终的水体分布图。利用了湖南省东北部不同成像模式的两景GF-3影像进行试验,在成像时间接近的光学影像中随机选择检验样点进行精度评价。实验结果表明,KI方法在GF-3水体提取应用中比Otsu方法具有更强的优势,剔除图像阴影区域后,自动化确定的阈值与目视解译阈值更加接近,通过MRF模型优化以后,实现了对水体信息的连贯提取,对图像噪声具有较强的抑制作用。本研究对水体目标的提取精度均达到了85%以上,实验结果精度优于基于光学影像的水体指数法,整个流程需要很少的人工经验参与,具有自动化程度强、检测精度高的优势。     

基于多特征的Gaofen-1海冰影像监督分割

冬季沿海地区的海冰检测工作对该地区居民的生产生活具有重要的指导意义,同时可以根据海冰变化检测气候变暖情况.海冰影像分割是海冰检测的基础.在众多海冰影像数据源中,Gaofen-1海冰影像因其丰富的光谱特征、较高的空间分辨率、简单的数据结构,在变化监测中具有重要应用价值.本文提出了一种基于红绿蓝3个光谱通道的灰度共生矩阵,...     

基于阈值分割与决策树的SAR影像水体信息提取

目前我国的GF-3 SAR数据可实现全天时全天候的对地观测,已服务于海洋、减灾、水利、气象等多个领域。改进了基于阈值分割法与决策树的GF-3 SAR影像水体信息提取方法,首先对GF-3 SAR影像进行基本处理;再采用KI二值化阈值分割法进行图像分割;然后通过构建知识决策树模型来提取水体信息,为了提高精度,采用GDEM数据进行地表建模,提取山体阴影;最后利用空间分析功能将地形建模阴影图与提取的水体范围进行匹配,去除山体阴影,进而获得水体信息的精确范围。通过混淆矩阵计算得到水体信息提取的总体精度为89.22%,Kappa系数为0.71,精度优于基于光学GF-2号影像的水体指数法提取结果。整个流程人工干预少,具有自动化更强、效率更高的优势。     

城市道路的多特征多核SVM提取方法

针对高分辨率遥感影像中城市道路提取的复杂性及SVM的分类性能,提出了一种城市道路的多特征多核SVM提取方法。首先利用FCM算法将原始影像粗分为建成区和非建成区两类,剔除非建成区;然后根据分水岭分割算法分割建成区并提取分割对象的光谱特征与空间特征,以全局核函数和局部核函数加权组合的方式构建多核SVM对建成区进行二次分类,去除建成区中的建筑物等非道路信息;最后利用数学形态学处理,获得最终的道路提取结果。试验结果表明：文中所提方法能够较精确地提取城市道路信息,分类精度高于单核SVM提取及其他对比方法。     

基于GMRF-SVM的高分辨率遥感影像目标区域划分方法

高分辨率遥感影像数据量大、细节丰富并呈现出一定的尺度依赖性,单一尺度遥感影像分割难以同时兼顾影像的宏观和微观特征,这成为制约遥感信息自动化提取技术发展的瓶颈之一。对此本文提出了基于特征的多尺度高分辨率遥感信息提取技术框架,并分析了其对于大尺度海量数据信息提取与目标识别工作具有的理论及实践意义。根据影像光谱或纹理等特征,提出采用GMRF-SVM方法在大尺度上进行分类的目标区域划分方法。从大尺度信息提取的角度来看,该方法综合了GMRF纹理分类和SVM少量样本模式识别的优势,便于先验知识的融合,无论从花费时间还是分类处理效果上,都远远优于直接采用GMRF进行分割所取得的效果,对于后面的信息提取和目标识别来说更具有实际意义。     

Crop classification from full-year fully-polarimetric L-band UAVSAR time-series using the Random Forest algorithm

Accurate and timely information on the distribution of crop types is vital to agricultural management, ecosystem services valuation and food security assessment. Synthetic Aperture Radar (SAR) systems have become increasingly popular in the field of crop monitoring and classification. However, the potential of time-series polarimetric SAR data has not been explored extensively, with several open scientific questions (e.g. the optimal combination of image dates for crop classification) that need to be answered. In this research, the usefulness of full year (both 2011 and 2014) L-band fully-polarimetric Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) data in crop classification was fully investigated over an agricultural region with a heterogeneous distribution of crop categories. In total, 11 crop classes including tree crops (almond and walnut), forage crops (grass, alfalfa, hay, and clover), a spring crop (winter wheat), and summer crops (corn, sunflower, tomato, and pepper), were discriminated using the Random Forest (RF) algorithm. The SAR input variables included raw linear polarization channels as well as polarimetric parameters derived from Cloude-Pottier (CP) and Freeman-Durden (FD) decompositions. Results showed clearly that the polarimetric parameters yielded much higher classification accuracies than linear polarizations. The combined use of all variables (linear polarizations and polarimetric parameters) produced the maximum overall accuracy of 90.50 % and 84.93 % for 2011 and 2014, respectively, with a significant increase of approximately 8 percentage points compared with linear polarizations alone. The variable importance provided by the RF illustrated that the polarimetric parameters had a far greater influence than linear polarizations, with the CP parameters being much more important than the FD parameters. The most important acquisitions were the images dated during the peak biomass stage (July and August) when the differences in structural characteristics between most crops were the largest. At the same time, the images in spring (April and May) and autumn (October) also contributed to the crop classification since they respectively provided unique information for discriminating fruit crops (almond and walnut) as well as summer crops (corn, sunflower, and tomato). As a result, the combined use of only four acquisitions (dated May, July, August, and October for 2011 and April, June, August, and October for 2014) was adequate to achieve a nearly-optimal overall accuracy. In light of the promising classification accuracies demonstrated in this research, it becomes increasingly viable to provide accurate and up-to-date crops inventories over large areas based solely on multitemporal polarimetric SAR.     

构建地块二维表征及CNN模型的作物遥感分类

本文旨在研究基于地块数据约束的深度学习模型的分类特征表示方法,以识别不同作物在不同时相上光谱差异从而对作物类型进行分类。通过Google Earth Engine平台获取作物生育期内全部Landsat 8影像,利用其质量评定波段完成研究区无云时相及区域上的地块统计,提取地块级别的各波段反射率均值按照时相顺序及波长进行排列,构建波谱、时相二维特征图作为该地块的抽象表示。通过构建相对最优的卷积神经网络CNN(Convolutional Neural Network)结构完成对特征图的分类,从而完成对地块的分类。构建CNN模型并不需要手工特征和预定义功能的需求,可完成提取特征并遵循端到端原则进行分类。将该模型的分类结果与其他最为常用机器学习分类器进行了比较,获得了优于常用遥感分类算法的分类精度。结果表明地块数据的加入可以有效的缩减计算规模并提供了准确的分类边界。所提出得方法在地块特征表示及作物分类中具有突出的应用潜力,应视为基于地块的多时相影像分类任务的优选方法。     

基于最优尺度和规则的高分辨率影像分类研究

高分辨率影像的广泛应用推进面向对象影像分析(OBIA)的发展,而分割作为面向对象分类的关键步骤,其尺度的选择直接关系到地物信息的提取。空间尺度是地物的固有属性,在合适的分割尺度下可以更好地挖掘地物信息。本文结合最大面积法和分割质量评价模型对张山营镇影像进行分割实验,先通过分析对象最大面积初步得到最优尺度范围,后结合分割质量评价模型以确定最优分割尺度层次。在此基础上,综合样本提取的光谱、纹理等特征进行规则训练,最终完成面向对象的土地覆被分类研究。结果显示:基于多层次最优尺度的规则分类方法获得更好的分类结果,其总体精度为88.8%,Kappa系数为0.861,而基于单一尺度的最邻近法总体精度81.4%,Kappa系数0.773,基于单一尺度的规则分类法总体精度为83.2%,Kappa系数为0.85。     

Evaluation of data fusion and image segmentation in earth observation based rapid mapping workflows

This paper is an exploratory study, which aimed to discover the synergies of data fusion and image segmentation in the context of EO-based rapid mapping workflows. Our approach pillared on the geographic object-based image analysis (GEOBIA) focusing on multiscale, internally-displaced persons’ (IDP) camp information extraction from very high spatial resolution (VHSR) images. We applied twelve pansharpening algorithms to two subsets of a GeoEye-1 image scene that was taken over a former war-induced ephemeral settlement in Sri Lanka. A multidimensional assessment was employed to benchmark pansharpening algorithms with respect to their spectral and spatial fidelity. The multiresolution segmentation (MRS) algorithm of the eCognition Developer software served as the key algorithm in the segmentation process. The first study site was used for comparing segmentation results produced from the twelve fused products at a series of scale, shape, and compactness settings of the MRS algorithm. The segmentation quality and optimum parameter settings of the MRS algorithm were estimated by using empirical discrepancy measures. Non-parametric statistical tests were used to compare the quality of image object candidates, which were derived from the twelve pansharpened products. A wall-to-wall classification was performed based on a support vector machine (SVM) classifier to classify image objects candidates of the fused images. The second site simulated a more realistic crisis information extraction scenario where the domain expertise is crucial in segmentation and classification. We compared segmentation and classification results of the original images (non-fused) and twelve fused images to understand the efficacy of data fusion. We have shown that the GEOBIA has the ability to create meaningful image objects during the segmentation process by compensating the fused image’s spectral distortions with the high-frequency information content that has been injected during fusion. Our findings further questioned the necessity of the data fusion step in rapid mapping context. Bypassing time-intensive data fusion helps to actuate EO-based rapid mapping workflows. We, however, emphasize the fact that data fusion is not limited to VHSR image data but expands over many different combinations of multi-date, multi-sensor EO-data. Thus, further research is needed to understand the synergies of data fusion and image segmentation with respect to multi-date, multi-sensor fusion scenarios and extrapolate our findings to other remote sensing application domains beyond EO-based crisis information retrieval.     

Efficient paddy field mapping using Landsat-8 imagery and object-based image analysis based on advanced fractel net evolution approach

This paper proposes an efficient paddy field mapping method using object-based image analysis and a bitemporal data set acquired by Landsat-8 Operational Land Imager. In the proposed approach, image segmentation is the first step and its quality has a serious impact on the accuracy of paddy field classification. In order to improve segmentation quality, a new segmentation algorithm based on a frequently used method, fractal net evolution approach, is developed, with improvement mainly in merging criteria. In order to automate the process of scale parameter determination, an unsupervised scale selection method is utilized to determine the optimal scale parameter for the proposed image segmentation approach. After segmentation, four types of object-based features including geometric, spectral, textural, and contextual information are extracted and input into the subsequent classification procedure. By using a random forest classifier, paddy fields and nonpaddy fields are separated. The proposed image segmentation method and the final classification result are both quantitatively evaluated. Our segmentation method outperformed two popular algorithms according to three supervised evaluation criteria. The classification result with overall accuracy of 91.00% and kappa statistic of 0.82 validated the effectiveness of the proposed framework. Further analysis on feature importance indicated that spectral features made the most contribution as compared to the other three types of object-based features.     

统计数据总量约束下全局优化阈值的冬小麦分布制图

大范围、长时间和高精度农作物空间分布基础农业科学数据的准确获取对资源、环境、生态、气候变化和国家粮食安全等问题研究具有重要现实意义和科学意义。本文针对传统阈值法农作物识别过程中阈值设置存在灵巧性差和自动化程度低等弱点,以中国粮食主产区黄淮海平原内河北省衡水市景县为典型实验区,首次将全局优化算法应用于阈值模型中阈值优化选取,开展了利用全局优化算法改进基于阈值检测的农作物分布制图方法创新研究。以冬小麦为研究对象,国产高分一号(GF-1)为主要遥感数据源,在作物面积统计数据为总量控制参考标准和全局参数优化的复合型混合演化算法SCE-UA (Shuffled Complex Evolution-University of Arizona)支持下,提出利用时序NDVI数据开展阈值模型阈值参数自动优化的冬小麦空间分布制图方法。最终,获得实验区冬小麦阈值模型最优参数,并利用优化后的阈值参数对冬小麦空间分布进行提取。通过地面验证表明,利用本研究所提方法获取的冬小麦识别结果分类精度均达到较高水平。其中冬小麦识别结果总量精度达到了99.99%,证明本研究所提阈值模型参数优化方法冬小麦提取分类结果总量控制效果良好;同时,与传统的阈值法、最大似然和支持向量机等分类方法相比,本研究所提阈值模型参数优化法区域冬小麦作物分类总体精度和Kappa系数分别都有所提高,其中,总体精度分别提高4.55%、2.43%和0.15%,Kappa系数分别提高0.12、0.06和0.01,这体现出SCE-UA全局优化算法对提高阈值模型冬小麦空间分布识别精度具有一定优势。以上研究结果证明了利用本研究所提基于作物面积统计数据总量控制以及SCE-UA全局优化算法支持下阈值模型参数优化作物分布制图方法的有效性和可行性,可获得高精度冬小麦作物空间分布制图结果,这对提高中国冬小麦空间分布制图精度和自动化水平具有一定意义,也可为农作物面积农业统计数据降尺度恢复重建和大范围区域作物空间分布制图研究提供一定技术参考。     

一种从SAR影像到光学影像的翻译方法

光学传感器在夜晚和云雨天气难以成像,合成孔径雷达（synthetic aperture radar,SAR）虽然能够全天时、全天候工作,但其成像难以理解,对此提出利用SAR影像翻译为光学影像的新思路来弥补二者的缺陷。给出了遥感影像翻译定义,提出一套包含图像理解、目标转换等环节的影像翻译技术流程。通过支持向量机分类、种子填充和基于样本的纹理合成算法等手段实现SAR影像典型目标向光学影像的转换与表达。最后,利用该方法实现了ENVISAT-ASAR转换为Landsat TM,ALOS PALSAR转换为GeoEye的两类影像翻译,并利用SAR影像翻译结果修补光学影像空缺。影像翻译和补缺实验证明了SAR影像翻译为光学影像的可行性和有效性。     

Estimation of Snow Cover Area,Snow Physical Properties and Glacier Classification in Parts of Western Himalayas Using C-Band SAR Data

Snow physical properties, snow cover and glacier facies are important parameters which are used to quantify snowpack characteristics, glacier mass balance and seasonal snow and glacier melt. This study has been done using C-band synthetic aperture radar (SAR) data of Indian radar imaging satellite, radar imaging satellite-1 (RISAT)-1, to estimate the seasonal snow cover and retrieve snow physical properties (snow wetness and snow density), and glacier radar zones or facies classification in parts of North West Himalaya (NWH), India. Additional SAR data used are of Radarsat-2 (RS-2) satellite, which was used for glacier facies classification of Smudra Tapu glacier in Himachal Pradesh. RISAT-1 based snow cover area (SCA) mapping, snow wetness and snow density retrieval and glacier facies classification have been done for the first time in NWH region. SAR-based inversion models were used for finding out wet and dry snow dielectric constant, dry and wet SCA, snow wetness and snow density. RISAT-1 medium resolution scan-SAR mode (MRS) in HV polarization was used for first time in NWH for deriving time series of SCA maps in Beas and Bhagirathi river basins for years 2013–2014. The SAR-based inversion models were implemented separately for RISAT-1 quad pol. FRS2, for wet snow and dry snow permittivity retrieval. Masks for layover and shadow were considered in estimating final snow parameters. The overall accuracy in terms of R² value comes out to be 0.74 for snow wetness and 0.72 for snow density based on the limited ground truth data for subset area of Manali sub-basin of Beas River up to Manali for winter of 2014. Accuracy for SCA was estimated to be 95 % when compared with optical remote sensing based SCA maps with error of ±10 %. The time series data of RISAT-1 MRS and hybrid data in RH/RV mode based decompositions were also used for glacier radar zones classification for Gangotri and Samudra Tapu glaciers. The various glaciers radar zones or facies such as debris covered glacier ice, clean or bare glacier ice radar zone, percolation/refreeze radar zone and wet snow, ice wall etc., were identified. The accuracy of classified maps was estimated using ground truth data collected during 2013 and 2014 glacier field work to Samudra Tapu and Gangotri glaciers and overall accuracy was found to be in range of 82–90 %. This information of various glacier radar zones can be utilized in marking firn line of glaciers, which can be helpful for glacier mass balance studies.     

基于双频多极化数据的农作物后向散射特性模拟分析及类型识别

以北京昌平地区为研究区域,获取了2007年该试验区C波段ENVISAT/ASAR数据和L波段ALOS/PALSAR数据,并提取了地物的后向散射系数。首先,利用MIMICS模型对该地区的春玉米、夏玉米和果木的后向散射特性进行模拟和分析;然后,将模拟结果同雷达实际观测数据进行对比;最后,利用不同作物之间的后向散射系数数值大小关系,建立分类二叉树,很好地区分了春玉米和夏玉米,总分类精度达86.66%。研究结果表明:双频多极化雷达数据能够提供有利于作物类型识别的多方面信息,对农作物遥感具有较大的优势和潜力。     

Evaluation of image fusion methods using PALSAR,RADARSAT-1 and SPOT images for land use/ land cover classification

This research aimed to explore the fusion of multispectral optical SPOT data with microwave L-band ALOS PALSAR and C-band RADARSAT-1 data for a detailed land use/cover mapping to find out the individual contributions of different wavelengths. Many fusion approaches have been implemented and analyzed for various applications using different remote sensing images. However, the fusion methods have conflict in the context of land use/cover (LULC) mapping using optical and synthetic aperture radar (SAR) images together. In this research two SAR images ALOS PALSAR and RADARSAT-1 were fused with SPOT data. Although, both SAR data were gathered in same polarization, and had same ground resolution, they differ in wavelengths. As different data fusion methods, intensity hue saturation (IHS), principal component analysis, discrete wavelet transformation, high pass frequency (HPF), and Ehlers, were performed and compared. For the quality analyses, visual interpretation was applied as a qualitative analysis, and spectral quality metrics of the fused images, such as correlation coefficient (CC) and universal image quality index (UIQI) were applied as a quantitative analysis. Furthermore, multispectral SPOT image and SAR fused images were classified with Maximum Likelihood Classification (MLC) method for the evaluation of their efficiencies. Ehlers gave the best score in the quality analysis and for the accuracy of LULC on LULC mapping of PALSAR and RADARSAT images. The results showed that the HPF method is in the second place with an increased thematic mapping accuracy. IHS had the worse results in all analyses. Overall, it is indicated that Ehlers method is a powerful technique to improve the LULC classification.     

高分辨率SAR图像中的桥梁识别方法研究

根据高分辨率SAR图像中桥梁目标的特点提出了一种新的桥梁识别算法,主要由水体分割、感兴趣区域确定和桥梁识别三个步骤实现。通过直方图均衡,方便了分割阈值的选取;通过数据采样,提高了识别的有效性;通过目标特征的抽取,保证了识别结果的准确性。识别过程在初级处理中,利用快速算法得到感兴趣区域;在后续处理中利用目标特征对该区域内的潜在目标进行识别,取得了良好的效果。     

噪声抑制的多极化SAR海冰图像分割

合成孔径雷达(SAR)海冰图像分割对全球气候研究和保证船舶航行安全具有重要意义。现有的基于区域的马尔可夫随机场(MRF)多极化SAR分割方法,由于受相干斑噪声影响,其区域划分不尽合理,不能有效完成分割。因此,提出一种噪声抑制的多极化SAR海冰图像分割算法,首先在极化总功率图上引入降低噪声的滤波算法,合理划分初始区域,其次考虑区域之间的差异度,从而实现多极化SAR海冰图像的准确分割。以RADARSAT-2和SIR-C获得的全极化海冰图像为实验数据进行验证,结果表明:和其他较先进算法相比,本文算法优势明显,既能高效保持图像连通性,又能增强图像的细节信息,具有更高的分割精度。     

面向对象与卷积神经网络模型的GF-6 WFV影像作物分类

GF-6WFV影像是中国首颗带有红边波段的中高分辨率8波段多光谱卫星的遥感影像,对于其影像及红边波段对作物分类影响的研究利用亟待展开.本文结合面向对象和深度学习提出一种适用于GF-6 WFV红边波段的卷积神经网络(RE-CNN)遥感影像作物分类方法.首先采用多尺度分割和ESP工具选择最佳分割参数完成影像分割,通过面向对...     

一种有效的MSTAR SAR图像分割方法

提出了一种有效的MSTAR SAR图像分割方法。该方法首先对待处理图像进行过分割操作,得到过分割图像区域,然后对过分割后的图像进行图像区域级和像素级的特征提取,得到用于表示图像的特征向量,接着对MSTAR SAR图像使用空间隐含狄利克雷分配模型(sLDA)和马尔科夫随机场(MRF)建立本文所提出的模型,得到能量泛函,最后运用Graph-Cut算法和Branch-and-Bound算法对能量泛函进行优化,得到最终的分割结果。通过使用MSTAR SAR图像进行分割实验比较,仿真结果表明了方法的有效性。