超像素在多极化SAR数据分类中的应用--以ALOS PALSAR为例

针对已提出的极化合成孔径雷达数据地物分类方法较难同时获得地物边界及相邻信息的问题,并为了减少图像处理的消耗时间,本文引入一种超像素生成算法--线性迭代聚类方法,对日本先进对地观测卫星多极化SAR数据进行地物分类研究。本文以四川省彭州市与什邡市交界地区为研究区,先采用Pauli分解生成RGB假彩色图像并进行滤波,再以此为基础使用线性迭代聚类方法生成超像素,最后用支持向量机分类方法,合理选取极化熵、各向异性度及平均散射角等极化特征组合在一起作为分类参数,对基于像素超像素的极化SAR图像的分类结果进行对比分析。使用超像素比其他基于像素的分类方法能够获得更好的结果,基于超像素分类的总体精度为95.23%,Kappa系数为92.58%。

Semi-Automatic Fracture Mapping Using Cellular Neural Networks Applied to ALOS PALSAR 2 Images of the Western Highlands of Cameroon

In Cameroon in general and in the Highlands of Cameroon in particular, there is no fracture map since its realization is not easy. The region’s harsh accessibility and climatic conditions make it difficult to carry out geological prospecting field missions that require large investments. This study proposes a semi-automatic lineament mapping approach to facilitate the elaboration of the fracture map in the West Cameroon Highlands. It uses neural networks in tandem with PCI Geomatica’s LINE algorithm to extract lineaments semi-automatically from an ALOS PALSAR 2 radar image. The cellular neural network algorithm of Lepage et al (2000) is implemented to enhance the pre-processed radar image. Then, the LINE module of Geomatica is applied to the enhanced image for the automatic extraction of lineaments. Finally, a control and a validation of the expert by spatial analysis allows elaborating the fracture map. The results obtained show that neural networks enhance and facilitate the identification of lineaments on the image. The resulting map contains more than 1800 fractures with major directions N20° - 30°, NS, N10° - 20°, N50° - 60°, N70° - 80°, N80° - 90°, N100° - 110°, N110° - 120° and N130° - 140° and N140° - 150°. It can be very useful for geological and hydrogeological studies, and especially to inform on the productivity of aquifers in this region of high agro-pastoral and mining interest for Cameroon and the Central African sub-region.

基于HJ1B和ALOS/PALSAR数据的森林地上生物量遥感估算

森林地上生物量的精确估算能够减小碳储量估算的不确定性。为了探寻一种有效地提高森林生物量估算精度的方法,探讨了基于遥感物理模型和经验统计模型估算山地森林地上生物量的方法。首先,基于Li-Strahler几何光学模型和多元前向模式(MFM)进行模型模拟,结合查找表算法(LUT)从多光谱图像HJ1B估算贺兰山研究区的森林地上生物量。其次,采用统计方法建立了2种回归模型:(1)多光谱图像HJ1B进行混合像元分解(SMA),并与雷达图像ALOS/PALSAR进行图像融合建立生物量回归模型;(2)雷达图像ALOS/PALSAR后向散射系数和实测生物量建立了生物量回归模型。用实测数据对3种算法估算结果进行精度验证。研究结果表明:采用几何光学模型和MFM算法估算的森林地上生物量精度最好(决定系数R2=0.61,均方根误差RMSE=8.33 t/hm2,P<0.001),其估算地上生物量与实测值一致性较好,估算生物量精度略优于SMA估算的精度(R2=0.60,RMSE=9.417 t/hm2);ALOS/PALSAR多元回归估算的精度最差(R2=0.39,RMSE=14.89 t/hm2)。由此可见,采用几何光学模型和混合像元分解SMA适合估算森林地上生物量,利用这2种方法进行森林地上生物量遥感监测研究具有一定的应用潜力。

综合后向散射特征与极化特征的L波段SAR数据岩石分类

以菲律宾民都洛岛为研究区域,选取ALOS PALSAR双极化数据(极化方式为HH和HV极化)作为数据源,通过提取后向散射系数(Sigma0 HH和Sigma0 HV)和极化分解参数(熵、角和反熵),使用最大似然分类方法实现研究区的岩石单元分类和填图。在加入了极化分解参数之后,总体精度由仅使用后向散射系数的36.706%提高到65.000%。海岸带沼泽和珊瑚礁的F1分数超过了0.80,辉长岩和Mansalay组的F1分数超过了0.75。引入极化特征后,岩石单元的边界被更好地提取,Mansalay组和Mindoro变质岩与其他岩石的可分性增强。3个极化分解参数弥补了多种岩石单元的后向散射系数难以区分的不足,显著提高了岩石单元的可分性。研究表明,L波段SAR数据的极化分解参数和后向散射系数相结合能提高植被覆盖区岩石单元的分类精度。

L波段合成孔径雷达影像反演涌浪参数

针对L波段合成孔径雷达图像尚无波浪反演算法的问题,提出了利用高级陆地观测卫星(ALOS)搭载的相控阵列L波段合成孔径雷达(PALSAR)在日本近海的图像发展的算法,反演得到了涌浪参数(包括波长和波向)。研究假设L波段SAR对涌浪的成像机制为线性的,通过图像处理得到了高质量的PALSAR图像谱,从此谱可计算得到谱峰值波长和波传播方向。此图像谱所固有的180°方向模糊,可利用涌浪只向近岸传播的特点解决。通过对反演和日本"全国港湾海洋波浪情报网"(NOWPHAS)现场观测的波参数的比较,波长的偏差是-10.5m,均方差是18.3m,相关系数是0.94;波向的偏差是-1.3°,均方差是15.5°,相关系数是0.94。上述数据表明,反演方法中将L波段PALSAR对涌浪的成像机制假设为线性是可行的。

Machine learning and geostatistical approaches for estimating aboveground biomass in Chinese subtropical forests

Background:Aboveground biomass(AGB)is a fundamental indicator of forest ecosystem productivity and health and hence plays an essential role in evaluating forest carbon reserves and supporting the development of targeted forest management plans.Methods:Here,we proposed a random forest/co-kriging framework that integrates the strengths of machine learning and geostatistical approaches to improve the mapping accuracies of AGB in northern Guangdong Province of China.We used Landsat time-series observations,Advanced Land Observing Satellite(ALOS)Phased Array L-band Synthetic Aperture Radar(PALSAR)data,and National Forest Inventory(NFI)plot measurements,to generate the forest AGB maps at three time points(1992,2002 and 2010)showing the spatio-temporal dynamics of AGB in the subtropical forests in Guangdong,China.Results:The proposed model was capable of mapping forest AGB using spectral,textural,topographical variables and the radar backscatter coefficients in an effective and reliable manner.The root mean square error of the plotlevel AGB validation was between 15.62 and 53.78 t∙ha^(−1),the mean absolute error ranged from 6.54 to 32.32 t∙ha^(−1),the bias ranged from−2.14 to 1.07 t∙ha^(−1),and the relative improvement over the random forest algorithm was between 3.8%and 17.7%.The largest coefficient of determination(0.81)and the smallest mean absolute error(6.54 t∙ha^(−1)were observed in the 1992 AGB map.The spectral saturation effect was minimized by adding the PALSAR data to the modeling variable set in 2010.By adding elevation as a covariable,the co-kriging outperformed the ordinary kriging method for the prediction of the AGB residuals,because co-kriging resulted in better interpolation results in the valleys and plains of the study area.Conclusions:Validation of the three AGB maps with an independent dataset indicated that the random forest/cokriging performed best for AGB prediction,followed by random forest coupled with ordinary kriging(random forest/ordinary kriging),and the random forest model.The proposed random forest/co-kriging framework provides an accurate and reliable method for AGB mapping in subtropical forest regions with complex topography.The resulting AGB maps are suitable for the targeted development of forest management actions to promote carbon sequestration and sustainable forest management in the context of climate change.

全极化SAR数据反演桥面高度

提出一种SAR数据检测桥面高度的反演方法。用几何射线描述桥梁散射机制,依据去取向理论和分类参数分析,得到SAR成像中桥梁结构目标单次、二次、三次散射的成像规律。通过SAR图像分类参数的聚类与细化、滤波和Hough变换直线检测算法,检测出单次、二次和三次散射回波所成位置线图像,进一步构成桥面高度的反演算法。用机载Pi-SAR数据反演了日本鸣门大桥的桥面高度和桥面宽度,并与实测数据对比。按照该方法,进一步采用星载ALOS-PALSAR数据检测中国东海大桥的桥面高度。反演方法是可行的。

基于多源遥感数据的森林生物量估算模型研究

探讨了利用光学遥感图像HJ1B和多极化L波段微波遥感数据ALOS/PALSAR建立森林生物量估算模型的方法,利用统计回归方法建立了4种模型:1利用雷达图像ALOS/PALSAR后向散射系数和实测生物量建立的生物量回归模型;2利用多光谱图像HJ1B进行混合像元分解(spectral mixture analysis,SMA)后的组分图像与雷达图像ALOS/PALSAR进行图像融合建立的生物量回归模型;3利用多光谱图像HJ1B进行混合像元分解后的组分图像与实测生物量建立的生物量回归模型;4利用HJ1B图像的NDVI指数与实测生物量建立的生物量回归模型。对4种模型估算的生物量进行了对比分析。结果表明:第2种方法融合后图像与森林地上生物量之间存在较好的定量关系,估算生物量与实测生物量一致性较好,估算生物量精度优于其他模型结果。利用光学和微波图像协同遥感能够有效地提高森林生物量估算的精度,但并非所有的融合方法都能提高生物量估算的精度。利用雷达和光学图混合像元分解法进行植被生态系统监测研究具有一定的应用潜力。

Properties of L-band differential InSAR for monitoring mining-induced subsidence in coalfield of Jining, Northern China

The properties and feasibility of L-band differential InSAR for detecting and monitoring mining-induced subsidence were systematically analyzed and demonstrated. The largest monitored subsidence gradient of 7.9×10-3 and magnitude of 91 cm were firstly derived by theoretical derivation. Then, the stronger phase maintaining capacity and weaker sensitivity to minor land subsidence compared with C-band DInSAR were illustrated by phase simulation of the actual mine subsidence. Finally, the data processing procedure of two-pass DInSAR was further refined to accurately observe subsidence of a coalfield of Jining in Northern China using 7 ALOS PALSAR images. The largest monitored subsidence magnitude of 39.22 cm and other properties were better investigated by testing results interpretation and subsidence analysis, and the absolute difference varying from 0.5 mm to 17.9 mm was obtained by comparison with leveling-measured subsidence. All of results show that L-band DInSAR technique can investigate the location, amount, area and other detailed subsidence information with relatively higher accuracy.

2009年慕士塔格峰山地冰川表面冬季运动分布数据集

慕士塔格峰位于帕米尔高原西部,是我国西部山地冰川的集中分布区之一,长期监测该地区冰川不仅有益于评估水资源状况,也有助于气候变化方面的研究。本文基于日本ALOS/PALSAR卫星分别于2009年1月14日和3月1日获取覆盖慕士塔格峰地区(37°48′18″N^38°35′14″N,74°42′45″E^75°41′50″E)的SAR数据,借助改进的像素跟踪算法,通过精确去除卫星轨道和传感器姿态差异带来的全局性位移和地形起伏导致的地形效应误差,得到了该地区山地冰川表面高精度运动分布场(GeoTIFF格式,32位浮点型)。其空间分辨率约为20 m。非冰川区残余运动的统计分析表明其总体精度约为0.5 m/46 day。冰川运动分布表明,该地区冰川运动主要呈现为积累区速度快,消融区和末端运动速度慢的特点,冰川运动整体上与地形存在一定的相关性,其中个别中小型冰川呈现出较强的活动性。本数据集可以作为该地区山地冰川运动的本底调查资料,为慕士塔格峰地区山地冰川运动研究提供基础数据支撑。另外,山地冰川运动高精度监测将有助于研究其动力学特征和预测冰川运动导致的地质灾害,同时也为我国冰川资源普查提供了一种有效途径。

Extracting Eco-hydrological Information of Inland Wetland from L-band Synthetic Aperture Radar Image in Honghe National Nature Reserve, Northeast China

Taking a typical inland wetland of Honghe National Nature Reserve (HNNR), Northeast China, as the study area, this paper studied the application of L-band Synthetic Aperture Radar (SAR) image in extracting eco-hydrological information of inland wetland. Landsat-5 TM and ALOS PALSAR HH backscatter images were first fused by using the wavelet-IHS method. Based on the fused image data, the classification method of support vector machines was used to map the wetland in the study area. The overall mapping accuracy is 77.5%. Then, the wet and dry aboveground biomass estimation models, including statistical models and a Rice Cloudy model, were established. Optimal parameters for the Rice Cloudy model were calculated in MATLAB by using the least squares method. Based on the validation results, it was found that the Rice Cloudy model produced higher accuracy for both wet and dry aboveground biomass estimation compared to the statistical models. Finally, subcanopy water boundary information was extracted from the HH backscatter image by threshold method. Compared to the actual water borderline result, the extracted result from L-band SAR image is reliable. In this paper, the HH-HV phase difference was proved to be valueless for extracting subcanopy water boundary information.

Phenology and classification of abandoned agricultural land based on ALOS-1 and 2 PALSAR multi-temporal measurements

Agricultural crop abandonment negatively impacts local economy and environment since land,as a resource for agriculture,is not optimally utilized.To take necessary actions to rehabilitate abandoned agricultural lands,the identification of the spatial distribution of these lands must be acknowledged.While optical images had previously illustrated potentials in the identification of agricultural land abandonment,tropical areas often suffer cloud coverage problem that limits the availability of the imageries.Therefore,this study was conducted to investigate the potential of ALOS-1 and 2(Advanced Land Observing Satellite-1 and 2)PALSAR(Phased Array L-band Synthetic Aperture Radar)images for the identification and classification of abandoned agricultural crop areas,namely paddy,rubber and oil palm fields.Distinct crop phenology for paddy and rubber was identified from ALOS-1 PALSAR;nonetheless,oil palm did not demonstrate any useful phenology for discriminating between the abandoned classes.The accuracy obtained for these abandoned lands of paddy,rubber and oil palm was 93.33%±0.06%,78%±2.32%and 63.33%±1.88%,respectively.This study confirmed that the understanding of crop phenology in relation to image date selection is essential to obtain high accuracy for classifying abandoned and non-abandoned agricultural crops.The finding also portrayed that PALSAR offers a huge advantage for application of vegetation in tropical areas.

稀疏轨道条件下SAR几何校正轨道拟合策略

针对SAR影像头文件中提供的轨道矢量分布较为稀疏时,使用低阶多项式拟合轨道参数会存在较大误差,从而导致基于距离-多普勒模型的星载SAR影像几何校正精度较差的问题,提出了一种稀疏轨道条件下实用的轨道拟合策略。通过对轨道拟合误差进行分析,综合考虑距离-多普勒模型的解算与现有几何校正资源,选择使用高阶插值方法生成卫星成像时刻处足够的加密点轨道状态矢量,再对局部的轨道进行二阶多项式拟合,利用拟合的轨道模型系数进行几何校正。利用实测GPS点数据进行了SAR几何校正精度验证,结果表明该策略能够明显改善ALOS2PALSAR影像几何校正精度,证明了该方法的有效性和良好的适用性。

基于特征跟踪方法的Amery冰架上游流速估算

冰川流速分析对于研究南极冰盖物质平衡、海平面上升和全球气候变化具有重要意义。以ALOS/PALSAR影像为数据源,采用SAR特征跟踪方法,结合DEM数据估算Amery冰架上游冰川流速。结果表明,Amery冰架上游主流线流速为540—720 m·a-1,冰川流速随海拔的降低逐渐减小。受基岩和两侧山体的影响,主流线流速大,越靠近两侧山体流速越小。这个结果与NASA 2000年利用SAR重复轨道干涉测量方法测定的流速接近。基于裸露岩石作为特征点的分析表明,该方法误差较小,获取的冰川流速比较可靠。

Simple Relationship Analysis between L-Band Backscattering Intensity and the Stand Characteristics of Sugi (<i>Cryptomeria japonica</i>) and Hinoki (<i>Chamaecyparis obtusa</i>) Trees

In this study, we have performed an analysis between the L-band backscattering intensity derived from the slope corrected ALOS PALSAR remote sensing data and the?in-situ?stand biophysical parameter of Sugi (Cryptomeria japonica) and Hinoki (Chamaecyparis obtusa) trees at the forests of Chiba Prefecture, Japan. Diameter at breast height (DBH), tree height, and stem volume were statistically compared with the slope corrected sigma naught backscattering in an empirical approach. It was found that the relationship between the backscattering and the stand characteristics was strongly dependent on species showing different trends between the Sugi and Hinoki trees.?The Hinoki trees showed an increasing backscattering with increasing parameters (higher DBH, higher Tree height and higher stem volume), as it was mentioned on various researches, while the Sugi tree showed and decreasing backscattering with increasing parameters. We?have also found for the Sugi trees that the backscattering is affected strongly by the number of stems. We have assumed that this is because of the characteristics of the Sugi trees which have high moisture content in the heartwood of the stem, compared with other tree species in Japan. The results pave the way to the possibility for estimating biophysical parameters within the forests of Japan by considering such trends and at highly rugged areas by using slope corrected imagery of the SAR data.

Using Bayesian multitemporal classification to monitor tropical forest cover changes in Kalimantan,Indonesia

Significant areas of native forest in Kalimantan,on the island of Borneo,have been cleared for the expansion of plantations of oil palm and rubber.In this study multisource remote sensing was used to develop a time series of land cover maps that distinguish native forest from plantations.Using a study area in east Kalimantan,Landsat images were combined with either ALOS PALSAR or Sentinel-1 images to map four land cover classes(native forest,oil palm plantation,rubber plantation,non-forest).Bayesian multitemporal classification was applied to increase map accuracy and maps were validated using a confusion matrix;final map overall accuracy was>90%.Over 18 years from 2000 to 2018 nearly half the native forests in the study area were converted to either non-forest or plantations of either rubber or oil palm,with the highest losses between 2015 and 2016.Trending upwards from 2008 large areas of degraded or cleared forests,mapped as non-forest,were converted to oil palm plantation.Conversion of native forests to plantation mainly occurred in lowland and wetland forest,while significant forest regrowth was detected in degraded peatland.These maps will help Indonesia with strategies and policies for balancing economic growth and conservation.