Prediction of the presence of topsoil nitrogen from spaceborne hyperspectral data

Conventional methods of soil nitrogen extraction are time consuming, expensive and tedious. Remote sensing and Geographical Information System technologies can be used for the rapid and efficient prediction of the presence of soil nitrogen. However, studies are limited by and large to fields of larger and homogeneous units. This research concentrates on the prediction of topsoil nitrogen from harvested, scattered and small-sized agricultural fields of India using hyperspectral data. Spaceborne hyperspectral Hyperion data are used for the prediction of the presence of nitrogen. Multivariate partial least square regression method was used to predict the presence of nitrogen from reflectance. Reflectance data were pretreated using moving average and Savitzky–Golay filters which resulted in moderate prediction of R² 0.65 and 0.63 for calibration and validation, respectively. It can be inferred that Hyperion data can be effectively used for the prediction of the presence of soil nitrogen with a moderate level of accuracy even in case of scattered fields and fields of sizes approximately equal to the spatial resolution of the satellite.     

Angle effects of vegetation indices and the influence on prediction of SPAD values in soybean and maize

To study the anisotropy of vegetation indices (VIs) and explore its influence on the retrieval accuracy of canopy soil-plant analyzer development (SPAD) value, the bidirectional reflectance distribution function (BRDF) models of soybean and maize are calculated from the multi-angle hyperspectral images acquired by UAV, respectively. According to the reflectance extracted from the BRDF model, the dependences of 16 commonly-used VIs on observation angles are analyzed, and the SPAD values of maize and soybean canopy are predicted by using the 16 VI values at different observation angles and their combinations as input parameters. The results show that the 16 VIs have different sensitivity to angle in the principal plane: green ratio vegetation index (GRVI), ratio vegetation index (RVI), red edge chlorophyll index (CIRE), and modified chlorophyll absorption in reflectance index/optimized soil-adjusted vegetation index (MCARI/OSAVI) are very sensitive to angles, among which MCARI/OSAVI of maize fluctuated the most (138.83 %); in contrast, the green optimal soil adjusted vegetation index (GOSAVI), normalized difference vegetation index (NDVI), and green normalized difference vegetation index (GNDVI) hardly change with the observation angles. In terms of SPAD prediction, the accuracy of different VI is different, the mean absolute error (MAE) showed that MCARI1 provided the highest accuracy of retrieval for soybean (MAE=1.617), while for maize it was MCARI/OSAVI (MAE=2.422). However, when using the same VI, there was no significant difference in the accuracy of the predicted results, whether the VI from different angles was used or the combination of multi-angles was used. The present results provide guiding significance and practical value for the retrieval of SPAD value in vegetation canopies and in-depth applications of multi-angular remote sensing.     

基于高光谱遥感反射比的太湖水体叶绿素a含量估算模型

旨在寻找叶绿素a的高光谱遥感敏感波段并建立其定量估算模型。通过对太湖水体的连续监测，获得了从2004年6月到8月3个月的太湖水体高光谱数据和水质化学分析数据。利用实测的高光谱数据分析计算太湖水体的离水辐亮度和遥感反射比；然后，通过相关分析寻找反演叶绿素a浓度的高光谱敏感波段，进而建立反演太湖水体叶绿素a浓度的高光谱遥感定量估算模型，并用相关数据对模型进行精度分析。研究发现，水体的遥感反射比光谱在719nm和725nm存在两个峰，其中719nm处的峰更明显且稳定。通过模型的对比分析，发现用这两个峰值处的遥感反射比参与建模可以提高叶绿素a的估算精度；并且认为由反射比比值变量R719／R670所建立的线性模型对叶绿素a浓度的估算精度最理想。     

针叶林冠层热点及多角度无人机遥感观测及特征分析

多角度遥感观测是研究植被冠层BRDF (Bidirectional Reflectance Distribution Function)特性的重要手段,但目前对森林冠层进行连续间隔采样的多角度遥感观测及数据较少,热点方向的观测尤为缺乏。本研究基于无人机多角度高光谱成像系统,在主平面上对针叶林冠层以等角度连续间隔采样进行多角度观测,获取了主平面上多角度(包括热点和暗点)高光谱影像,并将观测结果与四尺度几何光学模型模拟结果进行对比分析。多角度观测获取的植被冠层反射率呈现出典型的植被方向反射特征,后向大部分角度观测的冠层反射率高于前向,在热点处出现峰值,在暗点附近方向出现最低值,观测天顶角VZA (View Zenith Angle)较大时表现出明显的"碗边效应"。结果表明:(1)观测的针叶林冠层反射率及BRDF特性与四尺度模型模拟基本一致,但红光波段模拟的热点反射率稍低于观测,前向观测VZA较大时模拟与观测结果差异稍大;(2)冠层结构及叶片光学特性的差异会导致观测到的BRDF特征不同;(3)观测的针叶林冠层BRDF呈现明显的光谱效应,不同波段呈现的各向异性特性不同,红光波段各向异性最强,近红外波段最弱;(4) BRDF的光谱效应差异导致观测到的植被指数也表现出各向异性,NDVI (Normalized Difference Vegetation Index)、PRI (Photochemical Reflectance Index)和MTCI(MERIS Terrestrial Chlorophyll Index)在热点方向最低,EVI (Enhanced Vegetation Index)在热点方向最高。本研究中无人机多角度成像观测提供的角度和高光谱信息,尤其是热点和暗点信息,在地物识别及分类、植被冠层结构反演及碳循环关键参数获取等研究方面具有较好的应用前景,在其它地物反射或热辐射等方向性特性研究中也具有较大的潜力。     

The Extraction of Urban Surface Water from Hyperspectral Data Based on Spectral Indices

To prevent confusion between water and buildings in the extraction of urban surface water from hyperspectral data, we analyzed the spectra of shadows and water in hyperspectral images, and proposed an anti-shadow water extraction method. This method first uses the normalized difference vegetation index (NDVI) for initial water extraction, then uses the height of the reflectance peak at 588 nm to eliminate the shadow of buildings. The method was validated by two hyperspectral datacubes, which were obtained for Jiaxing City and Zhoushan City in Zhejiang Province, China. Compared to the common spectral indices used to extract a water body, such as the NDVI, normalized difference water index, hyperspectral difference water index, and index of water index, the proposed method could effectively eliminate the shadow of buildings. The commission error reduced from more than 40% to about 15%, and the Kappa coefficient was increased from 60 and 70% to over 80% for the two datacubes. This indicated that the proposed method can inhibit the shadow of buildings and does not have a regional dependence.     

Detecting and Downscaling Wet Areas on Boreal Landscapes

This letter presents an approach to classify wet areas from European Remote Sensing 2 (ERS-2) synthetic aperture radar (SAR)-, Landsat Thematic Mapper (TM)-, and Light Detection and Ranging (LiDAR)-derived terrain data and downscale the result from the coarse resolution of satellite images to finer resolutions needed for land managers. Using discrete wavelet transform (DWT) and support vector machines (SVM), the algorithm finds multiple relationships between the radar, optical, and terrain data and wet areas at different spatial scales. Decomposing and reconstructing processes are performed using a 2-D DWT (2D-DWT) and inverse 2D-DWT respectively. The underlying relationships between radar, optical, and terrain data and wet areas are learned by training an SVM at the coarse resolution of the wet-area map. The SVM is then applied on the predictors at a finer resolution to produce wet-area detailing images, which are needed to reconstruct a finer resolution wet-area map. The algorithm is applied to a boreal landscape in northern Alberta, Canada, characterized by many wet-area features including ephemeral and permanent streams and wetlands.     

Retrieving aboveground biomass of wetland Phragmites australis (common reed) using a combination of airborne discrete-return LiDAR and hyperspectral data

Wetland biomass is essential for monitoring the stability and productivity of wetland ecosystems. Conventional field methods to measure or estimate wetland biomass are accurate and reliable, but expensive, time consuming and labor intensive. This research explored the potential for estimating wetland reed biomass using a combination of airborne discrete-return Light Detection and Ranging (LiDAR) and hyperspectral data. To derive the optimal predictor variables of reed biomass, a range of LiDAR and hyperspectral metrics at different spatial scales were regressed against the field-observed biomasses. The results showed that the LiDAR-derived H_p99 (99th percentile of the LiDAR height) and hyperspectral-calculated modified soil-adjusted vegetation index (MSAVI) were the best metrics for estimating reed biomass using the single regression model. Although the LiDAR data yielded a higher estimation accuracy compared to the hyperspectral data, the combination of LiDAR and hyperspectral data produced a more accurate prediction model for reed biomass (R² = 0.648, RMSE = 167.546 g/m², RMSE_r = 20.71%) than LiDAR data alone. Thus, combining LiDAR data with hyperspectral data has a great potential for improving the accuracy of aboveground biomass estimation.     

基于高光谱数据和RBF神经网络方法的草地叶面积指数反演

基于中国农业科学院在呼伦贝尔草原实测的120组草地冠层光谱反射率及相应的叶面积指数(LAI)数据,在进行主成分分析(PCA)实现降维处理的基础上,利用径向基函数(radial basis function,RBF)神经网络方法对草地LAI进行了高光谱反演研究.PCA结果表明,前9个主成分的累积贡献率达到了99.782％,能包含原光谱数据的绝大部分信息.将120组LAI及相应的9个主成分样本数据随机分为校正集数据(90组)和预测集数据(30组),分别用于神经网络模型的建立和LAI的预测.所构建的神经网络模型的模拟结果表明,RBF神经网络模型对校正集样本的模拟准确率达到100％(RMSE =0.009 6,R2 =0.999);预测集样本的实测LAI和模拟LAI之间的均方误差和决定系数分别为0.218 6和0.839,取得了较好的模拟效果,有效提高了传统的多元线性回归方程(RMSE =0.416 5,R2=0.570)的计算精度.     

地表覆盖分类数据对区域森林叶面积指数反演的影响

以江西省吉安市为研究区,将5种全球地表覆盖分类数据(包括美国地质调查局(USGS)、马里兰大学(UMD)和波士顿大学(BU)生成的3套数据和欧洲生成的2套数据)以及由TM影像生成的区域地表覆盖分类数据,分别与MODIS1km反射率资料结合,利用基于4尺度几何光学模型的LAI反演方法生成研究区的LAI。在1km和4km两种尺度上将反演的LAI与TM资料生成的LAI进行比较,评价地表覆盖分类数据对LAI反演结果的影响。结果表明,TM和欧洲太空局的GLOBCOVER地表覆盖分类数据用于反演LAI的结果较好,在1km尺度上,反演的LAI与统计模型估算的TMLAI相关的R2分别为0.44和0.40,在4km尺度上的R2分别为0.57和0.54;其次为波士顿大学的MODIS地表覆盖分类数据,据其反演的LAI与TMLAI相关的R2在1km和4km尺度上分别为0.38和0.51;而马里兰大学的UMD和欧洲的GLC2000地表覆盖分类数据会导致反演的LAI存在较大误差,据其反演的LAI与TMLAI之间的一致性较差,在1km和4km两种尺度上平均偏低20%左右;LAI的反演结果对聚集度系数具有强的敏感性。该研究表明,为了提高区域/全球LAI反演精度,需要有高质量的地表覆盖分类数据。     

Estimating chlorophyll content of crops from hyperspectral data using a normalized area over reflectance curve (NAOC)

The Normalized Area Over reflectance Curve (NAOC) is proposed as a new index for remote sensing estimation of the leaf chlorophyll content of heterogeneous areas with different crops, different canopies and different types of bare soil. This index is based on the calculation of the area over the reflectance curve obtained by high spectral resolution reflectance measurements, determined, from the integral of the red–near-infrared interval, divided by the maximum reflectance in that spectral region. For this, use has been made of the experimental data of the SPARC campaigns, where in situ measurements were made of leaf chlorophyll content, LAI and fCOVER of 9 different crops – thus, yielding 300 different values with broad variability of these biophysical parameters. In addition, Proba/CHRIS hyperspectral images were obtained simultaneously to the ground measurements. By comparing the spectra of each pixel with its experimental leaf chlorophyll value, the NAOC was proven to exhibit a linear correlation to chlorophyll content. Calculating the correlation between these variables in the 600–800 nm interval, the best correlation was obtained by computing the integral of the spectral reflectance curve between 643 and 795 nm, which practically covers the spectral range of maximum chlorophyll absorption (at around 670 nm) and maximum leaf reflectance in the infrared (750–800 nm). Based on a Proba/CHRIS image, a chlorophyll map was generated using NAOC and compared with the land-use (crops classification) map. The method yielded a leaf chlorophyll content map of the study area, comprising a large heterogeneous zone. An analysis was made to determine whether the method also serves to estimate the total chlorophyll content of a canopy, multiplying the leaf chlorophyll content by the LAI. To validate the method, use was made of the data from another campaign ((SEN2FLEX), in which measurements were made of different biophysical parameters of 7 crops, and hyperspectral images were obtained with the CASI imaging radiometer from an aircraft. Applying the method to a CASI image, a map of leaf chlorophyll content was obtained, which on, establishing comparisons with the experimental data allowed us to estimate chlorophyll with a root mean square error of 4.2 μg/cm², similar or smaller than other methods but with the improvement of applicability to a large set of different crop types.     

小波变换在高光谱决策树分类中的应用研究

近几年来，离散小波变换在遥感图像压缩、消噪和融合中得到了广泛的应用。利用航空飞行的高光谱图像数据，先后进行了主成分分析及10种小波变换，并应用分类回归树对其进行分类处理。将小波变换的分类结果与主成分分析的结果及不同的小波变换方法之间进行了对比。结果表明，在样本数相同的条件下，小波变换的分类精度均高于主成分分析，其中Haar小波的分类精度最高；小波变换后的分类对样本数量的要求要小于主成分分析。在样本数足够的情况下，主成分分析数据压缩率要高于小波变换。但小波变换在压缩的情况下，仍保留了原光谱的大部分信息。而主成分分析只保留了原图像的方差而无法保留波形。     

Multiple spectral similarity metrics for surface materials identification using hyperspectral data

Modern hyperspectral imaging and non-imaging spectroradiometer has the capability to acquire high-resolution spectral reflectance data required for surface materials identification and mapping. Spectral similarity metrics, due to their mathematical simplicity and insensitiveness to the number of reference labelled spectra, have been increasingly used for material mapping by labelling reflectance spectra in hyperspectral data labelling. For a particular hyperspectral data set, the accuracy of spectral labelling depends considerably upon the degree of unambiguous spectral matching achieved by the spectral similarity metric used. In this work, we propose a new methodology for quantifying spectral similarity for hyperspectral data labelling for surface materials identification. Developed adopting the multiple classifier system architecture, the proposed methodology unifies into a single framework the differential performances of eight different spectral similarity metrics for the quantification of spectral matching for surface materials. The proposed methodology has been implemented on two types of hyperspectral data viz. image (airborne hyperspectral images) and non-image (library spectra) for numerous surface materials identification. Further, the performance of the proposed methodology has been compared with the support vector machines (SVM) approach, and with all the base spectral similarity metrics. The results indicate that, for the hyperspectral images, the performance of the proposed methodology is comparable with that of the SVM. For the library spectra, the proposed methodology shows a consistently higher (increase of about 30% when compared to SVM) classification accuracy. The proposed methodology has the potential to serve as a general library search method for materials identification using hyperspectral data.     

基于反射光谱预测土壤重金属元素含量的研究

本文利用实验室实测的土壤反射光谱以及铅、镉、汞等重金属元素数据,进行土壤重金属元素含量快速预测的可行性研究。本文利用偏最小二乘回归方法,研究了反射率(R)、一阶微分(FDR)、反射率倒数的对数(lg(1/R))和波段深度(BD)等对预测精度的影响,对这几种光谱指标预测土壤重金属含量的能力进行了分析和评价,同时分析了多光谱数据估算土壤重金属元素含量的可行性。结果表明,反射率倒数的对数lg(1/R)是估算土壤重金属元素含量最好的光谱指标,尤其是Cd和Pb,检验精度R超过0.82。有机质、铁锰氧化物和黏土矿物对土壤重金属元素的吸附是可见光—近红外—短波红外光谱估算其含量的机理。多光谱数据同样具有估算土壤重金属元素含量的能力,但实际数据则要考虑多种因素的影响。     

Development of S-NPP VIIRS global surface type classification map using support vector machines

With the launch of the Joint Polar Satellite System (JPSS)/Suomi National Polar-orbiting Partnership (S-NPP) satellite in October 2011, many of the terrestrial remote sensing products generated from Moderate Resolution Imaging Spectroradiometer (MODIS), such as the global land cover map, have been inherited and expanded into the JPSS/S-NPP mission using the new Visible Infrared Imaging Radiometer Suite (VIIRS) data. In this study, an improved algorithm including the use of a new classifier support vector machines (SVM) classifier was proposed to produce VIIRS surface type maps. In addition to the new classification algorithm, a new post-processing strategy involving the use of new ancillary data to refine the classification output is implemented. As a result, the new global International Geosphere-Biosphere Programme (IGBP) map based on the 2014 VIIRS surface reflectance data was generated with a 78.5 ± 0.6% overall classification accuracy. The new map was compared to a previously delivered VIIRS surface type map, and to the MODIS land cover product. Validation results including the error matrix, overall accuracy, and the user’s and producer’s accuracy suggest the new global surface type map provides similar classification accuracy compared to the old VIIRS surface type map, with higher accuracy achieved in agricultural types.     

Calibration of ASTER and ETM + imagery using empirical line method—A case study of north-east of Hajjah, Yemen

This study is aimed at using the Empirical Line Method (ELM) to eliminate atmospheric effects with respect to visible and near infrared bands of advanced spaceborne thermal emission and reflection radiometer (ASTER) and enhanced thematic mapper plus (ETM+) data. Two targets (Amran limestone as light target and quartz-biotite-sericite-graphite schists as dark target), which were widely exposed and easy to identify in the imagery were selected. The accuracy of the atmospheric correction method was evaluated from three targets (vegetation cover, Amran limestone and Akbra shale) of the surface reflectance. Analytical spectral devices (ASD) FieldSpec3 was used to measure the spectra of target samples. ETM+ data were less influenced by the atmospheric effect when compared to ASTER data. Normalized differences vegetation indices (NDVI) displayed good results with reflectance data when compared with digital number (DN) data because it is highly sensitive to ground truth reflectance (GTR). Most of the differences observed before and after calibration of satellite images (ASTER and ETM+) were absorbed in the SWIR region.      

Dimensionality Reduction and Classification of Hyperspectral Images Using Object-Based Image Analysis

Object-based image analysis (OBIA) has attained great importance for the delineation of landscape features, particularly with the accessibility to satellite images with high spatial resolution acquired by recent sensors. Statistical parametric classifiers have become ineffective mainly due to their assumption of normal distribution, vast increase in the dimensions of the data and availability of limited ground sample data. Despite pixel-based approaches, OBIA takes semantic information of extracted image objects into consideration, and thus provides more comprehensive image analysis. In this study, Indian Pines hyperspectral data set, which was recorded by the AVIRIS hyperspectral sensor, was used to analyse the effects of high dimensional data with limited ground reference data. To avoid the dimensionality curse, principal component analysis (PCA) and feature selection based on Jeffries–Matusita (JM) distance were utilized. First 19 principal components representing 98.5% of the image were selected using the PCA technique whilst 30 spectral bands of the image were determined using JM distance. Nearest neighbour (NN) and random forest (RF) classifiers were employed to test the performances of pixel- and object-based classification using conventional accuracy metrics. It was found that object-based approach outperformed the traditional pixel-based approach for all cases (up to 18% improvement). Also, the RF classifier produced significantly more accurate results (up to 10%) than the NN classifier.     

改进约束D-TIN的等高线陡坡区域识别

针对高线图中陡坡区域特征的自动识别问题,在应用Delaunay三角网与线性插值法生成等高线图的基础上,改进了基于最大角原则生成约束Delaunay三角网的算法,对等高线图进行了三角剖分,根据等高线之间约束三角形的几何参数,计算了等高线图坡面单元坡度,识别出坡度较高区域。进一步地,给出了一种扩张算法对相连陡坡单元进行划分形成陡坡区域,计算了各个陡坡区域中心坐标、面积与平均坡度。通过对12组有256个坐标高程值的数据构建约束D-TIN并生成三维地形图对识别结果进行了评价,识别正确率达0.903,平均识别时间为33ms。实验表明,对于不同数据生成的不同等级等高线图的识别结果均有较高的效率与准确率。     

结合Gram-Schmidt变换的高光谱影像谐波分析融合算法

针对高光谱影像谐波分析融合(HAF)算法在影像融合时不顾及地物光谱曲线整体反射率这一缺陷,提出了结合Gram-Schmidt变换的高光谱影像谐波分析融合(GSHAF)改进算法。GSHAF算法可在完全保留融合前后像元光谱曲线波形形态的基础上,将高光谱影像融合简化为各像元光谱曲线的谐波余相组成的二维影像与高空间分辨率影像之间的融合。它是在原始高光谱影像光谱曲线被谐波分解为谐波余项、振幅和相位后,首先将其谐波余项与高空间分辨率影像进行GS变换融合,这样便可有效地修正融合后像元光谱曲线的反射率特征,随后再利用该融合影像与谐波振幅、相位进行谐波逆变换,完成高光谱影像谐波融合。本文最后利用Hyperion高光谱遥感影像与ALI高空间分辨率影像对GSHAF算法进行可行性分析,再以HJ-1A等卫星数据对其进行普适性验证,试验结果表明,GSHAF算法不仅可以完全地保留光谱曲线波形形态,而且融合后影像的地物光谱曲线反射率更接近真实地物。     

Satellite mapping of Baltic Sea Secchi depth with multiple regression models

Secchi depth is a measure of water transparency. In the Baltic Sea region, Secchi depth maps are used to assess eutrophication and as input for habitat models. Due to their spatial and temporal coverage, satellite data would be the most suitable data source for such maps. But the Baltic Sea’s optical properties are so different from the open ocean that globally calibrated standard models suffer from large errors. Regional predictive models that take the Baltic Sea’s special optical properties into account are thus needed. This paper tests how accurately generalized linear models (GLMs) and generalized additive models (GAMs) with MODIS/Aqua and auxiliary data as inputs can predict Secchi depth at a regional scale. It uses cross-validation to test the prediction accuracy of hundreds of GAMs and GLMs with up to 5 input variables. A GAM with 3 input variables (chlorophyll a, remote sensing reflectance at 678 nm, and long-term mean salinity) made the most accurate predictions. Tested against field observations not used for model selection and calibration, the best model’s mean absolute error (MAE) for daily predictions was 1.07 m (22%), more than 50% lower than for other publicly available Baltic Sea Secchi depth maps. The MAE for predicting monthly averages was 0.86 m (15%). Thus, the proposed model selection process was able to find a regional model with good prediction accuracy. It could be useful to find predictive models for environmental variables other than Secchi depth, using data from other satellite sensors, and for other regions where non-standard remote sensing models are needed for prediction and mapping. Annual and monthly mean Secchi depth maps for 2003–2012 come with this paper as Supplementary materials.     

Estimating mineral content of indigenous browse species using laboratory spectroscopy and sentinel-2 imagery

Trees provide low-cost organic inputs, with the potential to improve livelihoods for rural communities. Understanding foliar nutrients of tree species is crucial for integration of trees into agroecosystems. The study explored nitrogen (N), phosphorus (P), potassium (K) and calcium (Ca) concentrations of nine browse species collected from the bushveld region of South Africa using wet analysis and laboratory spectroscopy in the region 400–2500 nm, along with partial least squares (PLS) regression. We further explore the relationship between canopy reflectance of Sentinel-2 image and foliar N, P, K & Ca. Laboratory spectroscopy was significant for N estimation, while satellite imagery also revealed useful information about the estimation of nitrogen at landscape level. Nitrogen was highly correlated with spectral reflectance (R² = 0.72, p < 0.05) for winter and (R² = 0.88, p < 0.05) for summer, whilst prediction of phosphorus potassium and calcium were considered not accurate enough to be of practical use. Modelling the relationship using Sentinel-2 data showed lower correlations for nitrogen (R² = 0.44, p < 0.05) and the other nutrients when compared to the dried samples. The findings indicate that there is potential to assess and monitor resource quality of indigenous trees using nitrogen as key indicator. This multi-level remote sensing approach has promise for providing rapid plant nutrient analyses at different scales.