Full polarimetric PALSAR-based land cover monitoring in Cambodia for implementation of REDD policies

Abstract

Forest cover monitoring plays an important role in the implementation of climate change mitigation policies such as Kyoto protocol and Reducing Emissions from Deforestation and Forest Degradation (REDD). In this study, we have monitored land cover using the PALSAR (Phased Array type L-band Synthetic Aperture Radar) full polarimetric data based on incoherent target decomposition. Supervised classification technique has been applied on Cloude–Pottier decomposition, Freeman–Durden three component, and Yamaguchi four component decomposition for accurate mapping of different types of land cover classes. Based on confusion matrix derived from the predicted and defined pixels, the evergreen and sparsely deciduous forests have shown high producer's accuracy by Freeman–Durden three component and Yamaguchi four component classifications. The overall accuracy of Maximum Likelihood Classification by Yamaguchi four component is 94.1% with 0.93 kappa coefficient as compared to the 90.3% with 0.88 kappa coefficient by Freeman–Durden three component and 89.7% with 0.88 kappa coefficient by Cloude–Pottier decomposition. High accuracy of classification in a forested area using full polarimetric PALSAR data may have been because of high penetration of L-band SAR. The content of this study could be useful for the forest cover mapping during cloudy days needed for proper implementation of REDD policies in Cambodia.     

The potential of linear discriminative Laplacian eigenmaps dimensionality reduction in polarimetric SAR classification for agricultural areas

In this paper, the linear discriminative Laplacian eigenmaps (LDLE) dimensionality reduction (DR) algorithm is introduced to C-band polarimetric synthetic aperture radar (PolSAR) agricultural classification. A collection of homogenous areas of the same crop class usually presents physical parameter variation, such as the biomass and soil moisture. Furthermore, the local incidence angle also impacts a lot on the same crop category when the vegetation layer is penetrable with C-band radar. We name this phenomenon as the “observed variation of the same category” (OVSC). The most common PolSAR features, e.g., the Freeman–Durden and Cloude–Pottier decompositions, show an inadequate performance with OVSC. In our research, more than 40 coherent and incoherent PolSAR decomposition models are stacked into the high-dimensionality feature cube to describe the various physical parameters. The LDLE algorithm is then performed on the observed feature cube, with the aim of simultaneously pushing the local samples of the same category closer to each other, as well as maximizing the distance between local samples of different categories in the learnt subspace. Finally, the classification result is obtained by nearest neighbor (NN) or Wishart classification in the reduced feature space. In the simulation experiment, eight crop blocks are picked to generate a test patch from the 1991 Airborne Synthetic Aperture Radar (AIRSAR) C-band fully polarimetric data from of Flevoland test site. Locality preserving projections (LPP) and principal component analysis (PCA) are then utilized to evaluate the DR results of the proposed method. The classification results show that LDLE can distinguish the influence of the physical parameters and achieve a 99% overall accuracy, which is better than LPP (97%), PCA (88%), NN (89%), and Wishart (88%). In the real data experiment, the Chinese Hailaer nationalized farm RadarSat2 PolSAR test set is used, and the classification accuracy is around 94%, which is again better than LPP (90%), PCA (88%), NN (89%), and Wishart (85%). Both experiments suggest that the LDLE algorithm is an effective way of relieving the OVSC phenomenon.     

Full year crop monitoring and separability assessment with fully-polarimetric L-band UAVSAR: A case study in the Sacramento Valley,California

Spatial and temporal information on plant and soil conditions is needed urgently for monitoring of crop productivity. Remote sensing has been considered as an effective means for crop growth monitoring due to its timely updating and complete coverage. In this paper, we explored the potential of L-band fully-polarimetric Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) data for crop monitoring and classification. The study site was located in the Sacramento Valley, in California where the cropping system is relatively diverse. Full season polarimetric signatures, as well as scattering mechanisms, for several crops, including almond, walnut, alfalfa, winter wheat, corn, sunflower, and tomato, were analyzed with linear polarizations (HH, HV, and VV) and polarimetric decomposition (Cloude–Pottier and Freeman–Durden) parameters, respectively. The separability amongst crop types was assessed across a full calendar year based on both linear polarizations and decomposition parameters. The unique structure-related polarimetric signature of each crop was provided by multitemporal UAVSAR data with a fine temporal resolution. Permanent tree crops (almond and walnut) and alfalfa demonstrated stable radar backscattering values across the growing season, whereas winter wheat and summer crops (corn, sunflower, and tomato) presented drastically different patterns, with rapid increase from the emergence stage to the peak biomass stage, followed by a significant decrease during the senescence stage. In general, the polarimetric signature was heterogeneous during June and October, while homogeneous during March-to-May and July-to-August. The scattering mechanisms depend heavily upon crop type and phenological stage. The primary scattering mechanism for tree crops was volume scattering (>40%), while surface scattering (>40%) dominated for alfalfa and winter wheat, although double-bounce scattering (>30%) was notable for alfalfa during March-to-September. Surface scattering was also dominant (>40%) for summer crops across the growing season except for sunflower and tomato during June and corn during July-to-October when volume scattering (>40%) was the primary scattering mechanism. Crops were better discriminated with decomposition parameters than with linear polarizations, and the greatest separability occurred during the peak biomass stage (July-August). All crop types were completely separable from the others when simultaneously using UAVSAR data spanning the whole growing season. The results demonstrate the feasibility of L-band SAR for crop monitoring and classification, without the need for optical data, and should serve as a guideline for future research.     

Characterization of forests and deforestation in Cambodia using ALOS/PALSAR observation

In this study, we have demonstrated the capability of full polarimetric ALOS/Phased Array L-band Synthetic Aperture Radar data for the characterization of the forests and deforestation in Cambodia, to support climate change mitigation policies of Reducing Emission from Deforestation and Forest Degradation (REDD). We have observed mean backscattering coefficient (σ°), entropy (H), alpha angle (α), anisotropy (A), pedestal height (PH), Radar Vegetation Index (RVI) and Freeman–Durden three-component decomposition parameters. The observations show that the forest types and deforested area are showing variable polarimetric and backscattering properties because of the structural difference. Evergreen forest is characterized by a high value of σ° HV (?12.96 dB) as compared with the deforested area (σ° HV=?22.2 dB). The value of polarimetric parameters such as entropy (0.93), RVI (0.91), PH (0.41) and Freeman–Durden volume scattering (0.43) is high for evergreen forest, whereas deforested area is characterized by the low values of entropy (0.36) and RVI (0.17). Based on these parameters, it is found that σ° HV, entropy, RVI and PH provide best results among other parameters.     

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.     

结合Freeman分解与子孔径散射特性的极化SAR图像分类

本文结合Freeman分解和子孔径分析,提出一种新的极化SAR图像分类算法。该方法首先利用子孔径分解,产生不同方位观察角度下的子孔径图像,再利用Freeman分解对各个子孔径图像提取三种散射机理成分的功率,平均后对类别进行细分,最后使用Wishart统计分类器对类别进行分类划分得到最终结果。该方法考虑了极化散射机理在不同方位观察角度下的变化,能够取得较好的分类效果,能够保存主要极化散射特性的纯度,同时还可以动态地设定分类类别数。最后利用EMISAR获取的极化SAR数据进行了仿真,验证了该方法的有效性。     

全卷积网络和条件随机场相结合的全极化SAR土地覆盖分类

针对经典全卷积网络(fully convolution network,FCN)分类精度低、效果差,以及传统的极化合成孔径雷达(PolSAR)土地覆盖分类方法未充分考虑地物散射特性的问题,提出了一种结合改进FCN和条件随机场(conditional random field,CRF)的全极化SAR土地覆盖分类算法。首先,利用Freeman分解和Pauli分解建模全极化SAR影像,同时提取各分解对应的散射特征,参考Freeman分解散射功率获取其主散射分量对应的主散射地物;同时,借鉴在图像分类领域中具有卓越表现的FCN-Vgg19-8s网络,考虑其高层卷积参数量大和低层卷积模型参数优化程度不足,通过在高层和中层分别构建多尺度卷积组和代价函数设计了FCN-MD-8s网络,保证对整体模型参数进行降维和优化;以Freeman分解散射机理特征为基准,采用级连式迁移学习结构,实现FCN-MD-8s网络的模型训练和测试;然后,根据主散射分量所对应的主散射地物,在各分量预测图中提取出主特征地物,得到分量地物分类结果,并将其进行叠加得到全局粗分类;最后,利用全连接CRF结合Pauli相干分解重建假彩色图,对全局粗分类进行全局像素类别转移获得细分类结果。通过对分类结果定性和定量分析,可知提出算法具有有效性和可行性。     

综合多特征的极化SAR图像随机森林分类算法

为抑制相干斑噪声对极化SAR图像分类结果的干扰,本文提出一种综合多特征的极化SAR图像随机森林分类方法。该方法首先利用简单线性迭代聚类(SLIC)算法生成超像素作为分类单元;然后,基于高维极化特征图像,利用训练好的随机森林模型,统计决策树的分类投票数,计算各超像素的类别概率;最后,利用超像素间的空间邻域特征,采用概率松弛算法(PLR)迭代修正超像素的类别后验概率,并依据最大后验概率(MAP)准则得到分类结果;实现综合利用超像素和空间邻域特征,降低相干斑噪声干扰的极化SAR图像分类方法。实验对比结果表明:本文方法能得有效抑制极化SAR图像中相干斑噪声的干扰,得到高精度且光滑连续的分类结果。     

顾及分类器参数的全极化SAR图像特征选择与分类

全极化SAR获取的信息量远多于传统SAR,但信息量的增加并不能确保分类精度的提高,如何有效进行特征选择至关重要。针对自适应特征选择问题,提出一种顾及分类器参数的特征选择和分类方法。该方法以支持向量数为评估依据,结合遗传算法进行特征选择,并同时对分类器参数进行寻优;最后利用优选的特征集和模型参数进行分类。为验证算法的有效性,利用两组全极化数据进行了监督分类实验。实验结果表明,提出方法降低了SVM分类器对自身参数的敏感性,而且能在较少特征个数下具备良好的泛化性能,分类精度优于未经过特征选择和参数优化的方法。     

Simulation analysis on the relationship between the leaf area index and polarimetric parameters of crops

Abstract

The leaf area index (LAI) is an important parameter to quantitatively describe the structure of vegetation and crops. Uncertainty in the relationship between the LAI and polarimetric parameters is the key problem for LAI estimation from polarimetric synthetic aperture radar (POLSAR) data. However, the existing POLSAR data have difficulties meeting the demand of the aforementioned research. This paper analyses the correlations between the LAI and the polarimetric parameters derived from Cloude and Freeman decompositions using simulated POLSAR data based on a coherent scattering model for maize and wheat. The results show: (1) The POLSAR data at C-band with a large incidence angle (40 degrees) are very suitable for finding the LAI for maize and wheat. (2) For maize there is a strong correlation between the scattering type angle and the LAI at C-band with a large incidence angle, and the coherency entropy, anisotropy, and the power of the double-bounce scattering power component also have significant correlations with the LAI. (3) For wheat at C-band with a high incidence angle, although the correlation coefficient is low, there is still a correlation between the entropy, anisotropy and LAI. Besides, the volume scattering is suitable for extracting the LAI for wheat at X-band.     

3维块匹配小波变换的极化SAR非局部均值滤波

极化合成孔径雷达(SAR)图像受相干斑噪声的影响,难以很好地保持结构特性,针对这个问题提出了一种采用3维块匹配小波变换的非局部均值滤波算法NL-3DWT(Nonlocal Filter based on 3-D Patch Matching Wavelet Transform)。该算法使用块匹配的3维非抽样小波变换对极化总功率图进行预滤波,在此基础上使用边界对齐窗提取结构相似像素,同时使用Sigma范围选择极化SAR数据的散射相似像素,共同构成相似像素集合;构建结构保持权重函数增大图像结构信息在块相似性度量时的权重,最终实现极化SAR图像结构保持的相干斑抑制。该算法增强了图像结构特征的表达,提高了结构相似像素选择的准确性,机载极化SAR数据实验结果表明,NL-3DWT算法能够在抑制相干斑噪声的同时,更有效地保持极化SAR图像的结构特性和极化散射特性。     

Scattering property based contextual PolSAR speckle filter

Reliability of the scattering model based polarimetric SAR (PolSAR) speckle filter depends upon the accurate decomposition and classification of the scattering mechanisms. This paper presents an improved scattering property based contextual speckle filter based upon an iterative classification of the scattering mechanisms. It applies a Cloude-Pottier eigenvalue-eigenvector decomposition and a fuzzy H/α classification to determine the scattering mechanisms on a pre-estimate of the coherency matrix. The H/α classification identifies pixels with homogeneous scattering properties. A coarse pixel selection rule groups pixels that are either single bounce, double bounce or volume scatterers. A fine pixel selection rule is applied to pixels within each canonical scattering mechanism. We filter the PolSAR data and depending on the type of image scene (urban or rural) use either the coarse or fine pixel selection rule. Iterative refinement of the Wishart H/α classification reduces the speckle in the PolSAR data. Effectiveness of this new filter is demonstrated by using both simulated and real PolSAR data. It is compared with the refined Lee filter, the scattering model based filter and the non-local means filter. The study concludes that the proposed filter compares favorably with other polarimetric speckle filters in preserving polarimetric information, point scatterers and subtle features in PolSAR data.     

基于四分量散射模型的多极化SAR图像分类

基于四分量散射模型提出了一种多极化SAR(synthetic aperture radar)图像非监督分类算法。与Freeman三分量散射模型不同,四分量散射模型在Freeman三分量的基础上增加了螺旋散射分量(helix),该分量反映了复杂地貌和不规则城市建筑的散射机理,可以用来处理复杂的场景图像。算法强调了初始分类的重要性,在初始分类中考虑了混合散射机制像素的存在,从而提高了分类结果的精确度。聚类过程中,采用由四个散射分量组成的特征向量进行迭代聚类。为了实现算法的完全非监督,利用特征向量给出了一种新的聚类终止准则。NASA/JPL实验室AIRSAR全极化数据分类实验结果表明,该算法具有较好的分类效果,并获得了较高的分类精度。     

Water-Body types identification in urban areas from radarsat-2 fully polarimetric SAR data

This paper presents a novel method for supervised water-body extraction and water-body types identification from Radarsat-2 fully polarimetric (FP) synthetic aperture radar (SAR) data in complex urban areas. First, supervised water-body extraction using the Wishart classifier is performed, and the false alarms that are formed in built-up areas are removed using morphological processing methods and spatial contextual information. Then, the support vector machine (SVM), the classification and regression tree (CART), TreeBagger (TB), and random forest (RF) classifiers are introduced for water-body types (rivers, lakes, ponds) identification. In SAR images, certain other objects that are misclassified as water are also considered in water-body types identification. Several shape and polarimetric features of each candidate water-body are used for identification. Radarsat-2 PolSAR data that were acquired over Suzhou city and Dongguan city in China are used to validate the effectiveness of the proposed method, and the experimental results are evaluated at both the object and pixel levels. We compared the water-body types classification results using only shape features and the combination of shape and polarimetric features, the experimental results show that the polarimetric features can eliminate the misclassifications from certain other objects like roads to water areas, and the increasement of classification accuracy embodies at both the object and pixel levels. The experimental results show that the proposed methods can achieve satisfactory accuracies at the object level [89.4% (Suzhou), 95.53% (Dongguan)] and the pixel level [96.22% (Suzhou), 97.95% (Dongguan)] for water-body types classification, respectively.     

In-season crop classification using elements of the Kennaugh matrix derived from polarimetric RADARSAT-2 SAR data

Accurate spatio-temporal classification of crops is of prime importance for in-season crop monitoring. Synthetic Aperture Radar (SAR) data provides diverse physical information about crop morphology. In the present work, we propose a day-wise and a time-series approach for crop classification using full-polarimetric SAR data. In this context, the 4 × 4 real Kennaugh matrix representation of a full-polarimetric SAR data is utilized, which can provide valuable information about various morphological and dielectric attributes of a scatterer. The elements of the Kennaugh matrix are used as the parameters for the classification of crop types using the random forest and the extreme gradient boosting classifiers.The time-series approach uses data patterns throughout the whole growth period, while the day-wise approach analyzes the PolSAR data from each acquisition into a single data stack for training and validation. The main advantage of this approach is the possibility of generating an intermediate crop map, whenever a SAR acquisition is available for any particular day. Besides, the day-wise approach has the least climatic influence as compared to the time series approach. However, as time-series data retains the crop growth signature in the entire growth cycle, the classification accuracy is usually higher than the day-wise data.Within the Joint Experiment for Crop Assessment and Monitoring (JECAM) initiative, in situ measurements collected over the Canadian and Indian test sites and C-band full-polarimetric RADARSAT-2 data are used for the training and validation of the classifiers. Besides, the sensitivity of the Kennaugh matrix elements to crop morphology is apparent in this study. The overall classification accuracies of 87.75% and 80.41% are achieved for the time-series data over the Indian and Canadian test sites, respectively. However, for the day-wise data, a ∼6% decrease in the overall accuracy is observed for both the classifiers.     

机载X波段极化干涉SAR数据非监督分类方法研究

针对经典极化分类算法在处理机载X波段SAR数据时将过多地物分为体散射类型,并且容易受噪声影响,分类结果存在大量误分现象的问题,通过对机载X波段SAR数据非监督分类方法的研究,提出将极化干涉信息用于机载X波段极化干涉SAR数据的分类。通过运用极化干涉数据进行目标分解得到参数A1和A2对数据进行初始分类,然后结合改进的Wishart最大似然分类算法来进行地物的自适应分类。实验结果表明,该方法能有效避免平地效应的影响,抗噪性好,能正确区分三种典型散射类型,分类效果明显优于极化分类效果。     

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.     

Land use classification from multitemporal Landsat imagery using the Yearly Land Cover Dynamics (YLCD) method

Several previous studies have shown that the inclusion of the LST (Land Surface Temperature) parameter to a NDVI (Normalized Difference Vegetation Index) based classification procedure is beneficial to classification accuracy. In this work, the Yearly Land Cover Dynamics (YLCD) approach, which is based on annual behavior of LST and NDVI, has been used to classify an agricultural area into crop types. To this end, a time series of Landsat-5 images for year 2009 of the Barrax (Spain) area has been processed: georeferenciation, destriping and atmospheric correction have been carried out to estimate NDVI and LST time series for year 2009, from which YLCD parameters were estimated. Then, a maximum likelihood classification was carried out on these parameters based on a training dataset obtained from a crop census. This classification has an accuracy of 87% (kappa = 0.85) when crops are subdivided in irrigated and non-irrigated fields, and when cereal crops are aggregated in a single crop, and performs better than a similar classification from Landsat bands only. These results show that a good crop differentiation can be obtained although detailed crop separation may be difficult between similar crops (barley, wheat and oat) due to similar annual NDVI and LST behavior. Therefore, the YLCD approach is suited for vegetation classification at local scale. As regards the assessment of the YLCD approach for classification at regional and global scale, it will be carried out in a further study.     

A statistical significance of differences in classification accuracy of crop types using different classification algorithms

Crop classification is needed to understand the physiological and climatic requirement of different crops. Kernel-based support vector machines, maximum likelihood and normalised difference vegetation index classification schemes are attempted to evaluate their performances towards crop classification. The linear imaging self-scanning (LISS-IV) multi-spectral sensor data was evaluated for the classification of crop types such as barley, wheat, lentil, mustard, pigeon pea, linseed, corn, pea, sugarcane and other crops and non-crop such as water, sand, built up, fallow land, sparse vegetation and dense vegetation. To determine the spectral separability among crop types, the M-statistic and Jeffries–Matusita (J–M) distance methods have been utilised. The results were statistically analysed and compared using Z-test and χ²-test. Statistical analysis showed that the accuracy results using SVMs with polynomial of degrees 5 and 6 were not significantly different and found better than the other classification algorithms.     

高精度作物分布图制作

中国自然条件复杂 ,农业种植结构多样 ,地块小而分散 ,利用遥感影像制作作物分布图的精度很难满足农业遥感估产的需求。该文利用目前最高分辨率的商用遥感卫星 (QuickBird)影像 ,采用面向对象的影像分析方法提取耕地种植地块图 ,结合详细的地面调查制作高精度的作物分布图 ,为农业遥感估产服务。