Combining QuickBird,LiDAR, and GIS topography indices to identify a single native tree species in a complex landscape using an object-based classification approach

There are now a wide range of techniques that can be combined for image analysis. These include the use of object-based classifications rather than pixel-based classifiers, the use of LiDAR to determine vegetation height and vertical structure, as well terrain variables such as topographic wetness index and slope that can be calculated using GIS. This research investigates the benefits of combining these techniques to identify individual tree species. A QuickBird image and low point density LiDAR data for a coastal region in New Zealand was used to examine the possibility of mapping Pohutukawa trees which are regarded as an iconic tree in New Zealand. The study area included a mix of buildings and vegetation types. After image and LiDAR preparation, single tree objects were identified using a range of techniques including: a threshold of above ground height to eliminate ground based objects; Normalised Difference Vegetation Index and elevation difference between the first and last return of LiDAR data to distinguish vegetation from buildings; geometric information to separate clusters of trees from single trees, and treetop identification and region growing techniques to separate tree clusters into single tree crowns. Important feature variables were identified using Random Forest, and the Support Vector Machine provided the classification. The combined techniques using LiDAR and spectral data produced an overall accuracy of 85.4% (Kappa 80.6%). Classification using just the spectral data produced an overall accuracy of 75.8% (Kappa 67.8%). The research findings demonstrate how the combining of LiDAR and spectral data improves classification for Pohutukawa trees.     

Sentinel-2和AW3D30相结合的湿地提取

为了实现大范围湿地动态监测,本文以辽河入海口附近的盘锦湿地为研究区,基于Sentinel2-L1C和AW3D30 DSM数据,在随机森林分类的基础上,结合地形数字特征和多边形形状特征对研究区进行湿地信息提取。通过人工目视解译对该分类方法进行精度验证,结果表明:该方法的自动化程度较高,能够在较少的人工干预下提取湿地覆盖范围。提取结果精度较高,制图精度和总体精度分别为91.04%和82.65%,Kappa系数为0.599 7,说明本文所采用的计算机分类方法与人工目视解译方法具有较好的一致性。     

Mapping wetland characteristics using temporally dense Sentinel-1 and Sentinel-2 data: A case study in the St. Lucia wetlands,South Africa

Wetlands have been determined as one of the most valuable ecosystems on Earth and are currently being lost at alarming rates. Large-scale monitoring of wetlands is of high importance, but also challenging. The Sentinel-1 and -2 satellite missions for the first time provide radar and optical data at high spatial and temporal detail, and with this a unique opportunity for more accurate wetland mapping from space arises. Recent studies already used Sentinel-1 and -2 data to map specific wetland types or characteristics, but for comprehensive wetland characterisations the potential of the data has not been researched yet. The aim of our research was to study the use of the high-resolution and temporally dense Sentinel-1 and -2 data for wetland mapping in multiple levels of characterisation. The use of the data was assessed by applying Random Forests for multiple classification levels including general wetland delineation, wetland vegetation types and surface water dynamics. The results for the St. Lucia wetlands in South Africa showed that combining Sentinel-1 and -2 led to significantly higher classification accuracies than for using the systems separately. Accuracies were relatively poor for classifications in high-vegetated wetlands, as subcanopy flooding could not be detected with Sentinel-1’s C-band sensors operating in VV/VH mode. When excluding high-vegetated areas, overall accuracies were reached of 88.5% for general wetland delineation, 90.7% for mapping wetland vegetation types and 87.1% for mapping surface water dynamics. Sentinel-2 was particularly of value for general wetland delineation, while Sentinel-1 showed more value for mapping wetland vegetation types. Overlaid maps of all classification levels obtained overall accuracies of 69.1% and 76.4% for classifying ten and seven wetland classes respectively.     

Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data

ABSTRACT

In recent years, the data science and remote sensing communities have started to align due to user-friendly programming tools, access to high-end consumer computing power, and the availability of free satellite data. In particular, publicly available data from the European Space Agency’s Sentinel missions have been used in various remote sensing applications. However, there is a lack of studies that utilize these data to assess the performance of machine learning algorithms in complex boreal landscapes. In this article, I compare the classification performance of four non-parametric algorithms: support vector machines (SVM), random forests (RF), extreme gradient boosting (Xgboost), and deep learning (DL). The study area chosen is a complex mixed-use landscape in south-central Sweden with eight land-cover and land-use (LCLU) classes. The satellite imagery used for the classification were multi-temporal scenes from Sentinel-2 covering spring, summer, autumn and winter conditions. Using stratified random sampling, each LCLU class was allocated 1477 samples, which were divided into training (70%) and evaluation (30%) subsets. Accuracy was assessed through metrics derived from an error matrix, but primarily overall accuracy was used in allocating algorithm hierarchy. A two-proportion Z-test was used to compare the proportions of correctly classified pixels of the algorithms and a McNemar’s chi-square test was used to compare class-wise predictions. The results show that the highest overall accuracy was produced by support vector machines (0.758 ± 0.017), closely followed by extreme gradient boosting (0.751 ± 0.017), random forests (0.739 ± 0.018), and finally deep learning (0.733 ± 0.0023). The Z-test comparison of classifiers showed that a third of algorithm pairings were statistically different. On a class-wise basis, McNemar’s test results showed that 62% of class-wise predictions were significant from one another at the 5% level or less. Variable importance metrics show that nearly half of the top twenty Sentinel-2 bands belonged to the red edge (25%) and shortwave infrared (23%) portions of the electromagnetic spectrum, and were dominated by scenes from spring (38%) and summer (40%). The results are discussed within the scope of recent studies involving machine learning and Sentinel-2 data and key knowledge gaps identified. The article concludes with recommendations for future research.     

Sentinel-2卫星落叶松林龄信息反演

林龄结构信息能够有效反映区域森林群落不同生长阶段的固碳能力,对于评估森林生态系统的健康状况具有重要意义。本研究以中国温带典型优势树种落叶松林为研究对象,分别选择其芽萌动期、展叶期和落叶期时段的Sentinel-2影像,采用多元线性回归（MLR）、随机森林（RF）、支持向量机回归（SVR）、前馈反向传播神经网络（BP）以及多元自适应回归样条（MARS）等5种方法依次构建落叶松林龄反演模型。通过相关性分析首先确定最佳遥感反演物候期,并在此基础上根据相关性差异筛选出5个最优特征变量用于模型反演,分别为冠层含水量（CWC）,归一化水体指数（NDWI）,叶面积指数（LAI）,光合有效辐射吸收率（FAPAR）和植被覆盖度（FVC）。研究结果表明,展叶期为落叶松林最佳遥感反演物候期。除植被衰减指数（PSRI）以及落叶期的NDVI、RVI外,落叶松林龄与各指标之间均呈负相关关系,其中与冠层含水量（CWC）的相关性最高,pearson相关系数达到-0.74（p<0.01）。此外,不同模型反演结果表明,随机森林模型（RF）为最佳落叶松林龄估测模型,其平均决定系数R²和平均均方根误差RMSE分别为0.89和2.91 a;多元线性回归模型（MLR）的林龄估测结果最差,其平均决定系数R²和平均均方根误差RMSE仅为0.57和5.69 a,非线性模型能更好的解释林龄与建模变量之间的关系。     

High-resolution mapping of forest canopy height using machine learning by coupling ICESat-2 LiDAR with Sentinel-1, Sentinel-2 and Landsat-8 data

Forest canopy height is an important indicator of forest carbon storage, productivity, and biodiversity. The present study showed the first attempt to develop a machine-learning workflow to map the spatial pattern of the forest canopy height in a mountainous region in the northeast China by coupling the recently available canopy height (H_canopy) footprint product from ICESat-2 with the Sentinel-1 and Sentinel-2 satellite data. The ICESat-2 H_canopy was initially validated by the high-resolution canopy height from airborne LiDAR data at different spatial scales. Performance comparisons were conducted between two machine-learning models – deep learning (DL) model and random forest (RF) model, and between the Sentinel and Landsat-8 satellites. Results showed that the ICESat-2 H_canopy showed the highest correlation with the airborne LiDAR canopy height at a spatial scale of 250 m with a Pearson’s correlation coefficient (R) of 0.82 and a mean bias of -1.46 m, providing important evidence on the reliability of the ICESat-2 vegetation height product from the case in China’s forest. Both DL and RF models obtained satisfactory accuracy on the upscaling of ICESat-2 H_canopy assisted by Sentinel satellite co-variables with an R-value between the observed and predicted H_canopy equalling 0.78 and 0.68, respectively. Compared to Sentinel satellites, Landsat-8 showed relatively weaker performance in H_canopy prediction, suggesting that the addition of the backscattering coefficients from Sentinel-1 and the red-edge related variables from Sentinel-2 could positively contribute to the prediction of forest canopy height. To our knowledge, few studies have demonstrated large-scale vegetation height mapping in a resolution ≤ 250 m based on the newly available satellites (ICESat-2, Sentinel-1 and Sentinel-2) and DL regression model, particularly in the forest areas in China. Thus, the present work provided a timely and important supplementary to the applications of these new earth observation tools.     

Integration of SAR-derived river inundation areas,high-precision topographic data and a river flow model toward near real-time flood management

Since several space-borne synthetic aperture radar (SAR) instruments providing high spatial resolutions and multi-polarisation capabilities will be mounted on satellites to be launched from 2006 onwards, radar imagery promises to become an indispensable asset for many environmental monitoring applications. Due to its all weather, day and night capabilities, SAR imagery presents obvious advantages over optical instruments, especially in flood management applications. To date, however, the coarse spatial resolution of available SAR datasets restricts the information that can be reliably extracted and processing techniques tend to be limited to binary floodplain segmentation into ‘flooded’ and ‘non flooded’ areas. It is the purpose of this paper to further improve the exploitation of SAR images in hydraulic modelling and near real-time crisis management by means of developing image processing methodologies that allow for the extraction of water levels at any point of the floodplain. As high-precision digital elevation models (DEM) produced, for instance, from airborne laser scanning become more readily available, methods can be exploited that combine SAR-derived flood extent maps and precise topographic data for retrieving water depth maps. In a case study of a well-documented flood event in January 2003 on the River Alzette, Grand Duchy of Luxembourg, a root mean squared error (R.M.S.E.) of 41 cm was obtained by comparing the SAR-derived water heights with surveyed high water marks that were collected during image acquisition. Water levels that were computed by a previously calibrated hydraulic model also suggest that the water surface profiles provided by the combined use of topographic data and SAR accurately reflect the true water line. The extraction of flooded areas within vegetated areas further demonstrates the usefulness of the proposed methodology.     

利用Sentinel-2A多光谱成像仪与Landsat8陆地成像仪影像进行普陀山岛植被分类效果比较

卫星遥感技术可用于海岛资源调查。Sentinel-2A与Landsat 8两颗卫星都可免费提供空间分辨率较高的多光谱遥感影像,在海岛调查中的应用潜力较大。本文以浙江舟山普陀山岛为例开展了针对这两种影像在海岛植被分类中的应用效果的研究,分别利用Sentinel-2A多光谱成像仪（MSI）和Landsat 8陆地成像仪（OLI）影像基于最大似然法分类获得了该岛阔叶林、针阔混交林、针叶林、灌丛、草丛等植被及其他地物的分布情况,并进行了精度检验,结果表明MSI的总体分类精度略高于OLI。     

机载LiDAR和高光谱融合实现温带天然林树种识别

将机载LiDAR(Light Detection and Ranging)与高光谱CASI(Compact Airborne Spectrographic Imager)数据融合,充分利用垂直结构信息和光谱信息进行温带森林树种分类,并与仅用高光谱数据的分类结果相比较,评估融合数据的树种分类能力。结合样地实测数据,首先用LiDAR获得的3维垂直结构信息对CASI影像上的林间空隙进行掩膜,提取林木冠层子集;然后对冠层子集分层掩膜,利用光谱曲线的一阶微分及曲线匹配技术,实现各树种训练样本的自动提取;利用SVM分类器对两种数据分类并比较精度。结果表明,融合数据的树种分类总体精度和Kappa系数(83.88%,0.80)优于仅使用CASI数据(76.71%、0.71),优势树种的制图精度为78.43%—89.22%,用户精度为75.15%—95.65%,整体也优于仅使用CASI的制图精度(68.51%—84.69%)和用户精度(63.34%—95.45%)。结果表明,机载LiDAR与CASI基于像元的融合对温带森林树种识别的精度较仅高光谱数据有较大提高。     

Assessing the suitability of data from Sentinel-1A and 2A for crop classification

Sentinel-1A C-SAR and Sentinel-2A MultiSpectral Instrument (MSI) provide data applicable to the remote identification of crop type. In this study, six crop types (beans, beetroot, grass, maize, potato, and winter wheat) were identified using five C-SAR images and one MSI image acquired during the 2016 growing season. To assess the potential for accurate crop classification with existing supervised learning models, the four different approaches namely kernel-based extreme learning machine (KELM), multilayer feedforward neural networks, random forests, and support vector machine were compared. Algorithm hyperparameters were tuned using Bayesian optimization. Overall, KELM yielded the highest performance, achieving an overall classification accuracy of 96.8%. Evaluation of the sensitivity of classification models and relative importance of data types using data-based sensitivity analysis showed that the set of VV polarization data acquired on 24 July (Sentinel-1A) and band 4 data (Sentinel-2A) had the greatest potential for use in crop classification.     

Evaluation of different machine learning methods for land cover mapping of a Mediterranean area using multi-seasonal Landsat images and Digital Terrain Models

Land cover monitoring using digital Earth data requires robust classification methods that allow the accurate mapping of complex land cover categories. This paper discusses the crucial issues related to the application of different up-to-date machine learning classifiers: classification trees (CT), artificial neural networks (ANN), support vector machines (SVM) and random forest (RF). The analysis of the statistical significance of the differences between the performance of these algorithms, as well as sensitivity to data set size reduction and noise were also analysed. Landsat-5 Thematic Mapper data captured in European spring and summer were used with auxiliary variables derived from a digital terrain model to classify 14 different land cover categories in south Spain. Overall, statistically similar accuracies of over 91% were obtained for ANN, SVM and RF. However, the findings of this study show differences in the accuracy of the classifiers, being RF the most accurate classifier with a very simple parameterization. SVM, followed by RF, was the most robust classifier to noise and data reduction. Significant differences in their performances were only reached for thresholds of noise and data reduction greater than 20% (noise, SVM) and 25% (noise, RF), and 80% (reduction, SVM) and 50% (reduction, RF), respectively.     

A GEOBIA framework for the implementation of national and international forest definitions using very high spatial resolution optical satellite data

In recent decades, there is an increasing need for harmonised and accurate information on the status and extent of forests. However, delineating the extent of forest areas is a complex task, since the existence of more than 100 definitions of forest worldwide causes considerable discrepancies in forested area estimates. The aim of this work was to examine the potential of geographic object based image analysis (GEOBIA) and very high spatial resolution imagery to discriminate forest areas following two different definitions of forest in northern Greece. In particular, we examined the definition of forest under the Greek law as well as the United Nations Food and Agricultural Organisation definition. Our findings suggest that the developed GEOBIA approach not only performed remarkably well for the discrimination of forest areas but also allowed to estimate rapidly and reliably forest extents when the two aforementioned forest definitions were employed.     

耦合WorldView-2和珠海一号影像的红树林物种分布

红树林物种遥感分类受影像分辨率、波段信息、分类策略、影像特征选择方法等因素影响。现有的红树林物种遥感分类研究大多关注于分类精度的比较,较少报道商业高分与国产高光谱卫星的结合在物种分类的应用。基于湛江高桥红树林保护区的高分（WorldView-2）和高光谱（珠海一号）卫星影像,本文旨在利用随机森林分类方法,比较不同特征选择算法（极端梯度提升（XGBoost）、极端随机树（ERT）、连续投影算法（SPA））和不同空间分辨率（WorldView-2影像从0.5 m的分辨率重采样为1 m、2 m、4 m、8 m、10 m）对红树林物种分类精度的影响,并耦合WorldView-2与珠海一号影像研究红树林物种空间分布格局及影响因素。每种空间分辨率的WorldView-2影像提取248个特征（52个光谱特征及196个纹理特征）,珠海一号影像提取117个光谱特征。结果表明,XGBoost方法优于ERT和SPA,在影像特征选择中具有巨大优势;与其他空间分辨率相比,2 m分辨率的WorldView-2影像具有更高的分类精度;耦合WorldView-2和珠海一号影像的分类精度（分辨率：2 m,总体精度：88.98%,kappa系数：0.846）高于单一的WorldView-2影像（分辨率：2 m,总体精度：83.47%,kappa系数：0.768）和珠海一号影像（分辨率：10 m,总体精度：78.50%,kappa系数：0.703）;另外,林窗、高程、离岸距离一定程度上影响红树林物种分布格局。本文表明WorldView-2与珠海一号高光谱影像的结合在景观及区域尺度上精确提取红树林物种信息具有巨大潜力,有利于红树林生物多样性保护及科学管理。     

Study of coastal wetland classification based on decision rules using ALOS AVNIR-2 images and ancillary geospatial data

Wetlands play irreplaceable key roles in ecological and environmental procedures. To make effective conservation and management, it is essential to understand the wetlands’ distribution and changes. In this study, an approach based on decision rules algorithm in conjunction with maximum likelihood classification is proposed for coastal wetland mapping using multi-temporal remotely sensed imagery and ancillary geospatial data. As a case study, Multi-temporal Advanced Visible and Near Infrared Radiometer type 2 images acquired by Japanese Advanced Land Observation Satellite are analysed to investigate the seasonal change pattern of coastal wetlands in Washington State, USA. Geospatial data, including Digital Elevation Model and spatial neighbourhood knowledge, are further integrated to characterize wetland features and discriminate classes within a certain elevation ranges. The final result is a refined coastal wetland map with 15 land cover categories. Preliminary evaluation of the final result shows that the proposed approach is effective in coastal wetland mapping.     

Estimating aboveground biomass of the mangrove forests on northeast Hainan Island in China using an upscaling method from field plots,UAV-LiDAR data and Sentinel-2 imagery

The mangrove forests of northeast Hainan Island are the most species diverse forests in China and consist of the Dongzhai National Nature Reserve and the Qinglan Provincial Nature Reserve. The former reserve is the first Chinese national nature reserve for mangroves and the latter has the most abundant mangrove species in China. However, to date the aboveground ground biomass (AGB) of this mangrove region has not been quantified due to the high species diversity and the difficulty of extensive field sampling in mangrove habitat. Although three-dimensional point clouds can capture the forest vertical structure, their application to large areas is hindered by the logistics, costs and data volumes involved. To fill the gap and address this issue, this study proposed a novel upscaling method for mangrove AGB estimation using field plots, UAV-LiDAR strip data and Sentinel-2 imagery (named G∼LiDAR∼S2 model) based on a point-line-polygon framework. In this model, the partial-coverage UAV-LiDAR data were used as a linear bridge to link ground measurements to the wall-to-wall coverage Sentinel-2 data. The results showed that northeast Hainan Island has a total mangrove AGB of 312,806.29 Mg with a mean AGB of 119.26 Mg ha⁻¹. The results also indicated that at the regional scale, the proposed UAV-LiDAR linear bridge method (i.e., G∼LiDAR∼S2 model) performed better than the traditional approach, which directly relates field plots to Sentinel-2 data (named the G∼S2 model) (R² = 0.62 > 0.52, RMSE = 50.36 Mg ha⁻¹<56.63 Mg ha⁻¹). Through a trend extrapolation method, this study inferred that the G∼LiDAR∼S2 model could decrease the number of field samples required by approximately 37% in comparison with those required by the G∼S2 model in the study area. Regarding the UAV-LiDAR sampling intensity, compared with the original number of LiDAR plots, 20% of original linear bridges could produce an acceptable accuracy (R² = 0.62, RMSE = 51.03 Mg ha⁻¹). Consequently, this study presents the first investigation of AGB for the mangrove forests on northeast Hainan Island in China and verifies the feasibility of using this mangrove AGB upscaling method for diverse mangrove forests.     

Estimation of aboveground biomass and carbon in a tropical rain forest in Gabon using remote sensing and GPS data

The knowledge of biomass stocks in tropical forests is critical for climate change and ecosystem services studies. This research was conducted in a tropical rain forest located near the city of Libreville (the capital of Gabon), in the Akanda Peninsula. The forest cover was stratified in terms of mature, secondary and mangrove forests using Landsat-ETM data. A field inventory was conducted to measure the required basic forest parameters and estimate the aboveground biomass (AGB) and carbon over the different forest classes. The Shuttle Radar Topography Mission (SRTM) data were used in combination with ground-based GPS measurements to derive forest heights. Finally, the relationships between the estimated heights and AGB were established and validated. Highest biomass stocks were found in the mature stands (223 ± 37 MgC/ha), followed by the secondary forests (116 ± 17 MgC/ha) and finally the mangrove forests (36 ± 19 MgC/ha). Strong relationships were found between AGB and forest heights (R² > 0.85).     

Development of a modified neural network-based land cover classification system using automated data selector and multiresolution remotely sensed data

Integrating multiple images with artificial neural networks (ANN) improves classification accuracy. ANN performance is sensitive to training datasets. Complexity and errors compound when merging multiple data, pointing to needs for new techniques. Kohonen's self-organizing mapping (KSOM) neural network was adapted as an automated data selector (ADS) to replace manual training data processes. The multilayer perceptron (MLP) network was then trained using automatically extracted datasets and used for classification. Two hypotheses were tested: ADS adapted from the KSOM network provides adequate and reliable training datasets, improving MLP classification performance; and fusion of Landsat thematic mapper (TM) and SPOT images using the modified ANN approach increases accuracy. ADS adapted from the KSOM network improved training data quality and increased classification accuracy and efficiency. Fusion of compatible multiple data can improve performance if appropriate training datasets are collected. This proved to be a viable classification scheme particularly where acquiring sufficient and reliable training datasets is difficult.