高光谱与高空间分辨率遥感数据的亚热带森林生物量反演

精确估算森林生物量对全球碳平衡以及气候变化的研究有重要意义。以亚热带天然次生林为研究对象,借助地面实测样地数据,通过对机载LiCHy(LiDAR,CCD and Hyperspectral)传感器同时获取的高光谱和高空间分辨率数据进行信息提取和数据融合,建模反演森林生物量。首先通过面向对象分割方法进行单木冠幅提取,然后融合从高光谱数据提取的光谱特征变量和从高空间分辨率数据提取的单木冠幅统计变量,构建多元回归模型估算地上、地下生物量,最后利用地面实测生物量经交叉验证评价模型精度。结果表明,综合模型的精度(R~2为0.54—0.62)高于高光谱模型(R~2为0.48—0.57);在高光谱模型中地上生物量模型精度(R~2为0.57)高于地下生物量模型(R~2为0.48);在综合模型中地上生物量模型精度(R~2为0.62)同样高于地下生物量模型(R~2为0.54)。交叉验证结果表明,与仅使用高光谱数据(单一数据源)相比,通过集成高光谱和高空间分辨率数据的生物量反演效果有所提升,可以更加有效地估算亚热带森林生物量。     

利用高分辨率森林覆盖影像实现高山林线的自动提取

快速准确地提取高山林线对标定全球气候变化、科学管理森林资源具有重要意义。秦岭高山林线位于高海拔的生态交错带,垂直带谱分布明显,为本文提供了理想的试验区域。遥感技术具有重访周期短、观测范围大和不受地理环境等条件限制的特点,克服了传统实地调查方法效率低、成本高等缺点。本文基于全球首套30 m空间分辨率森林覆盖数据,结合数字高程模型以及秦岭山脉分布数据,提出一种基于遥感的自动提取高山林线的算法;结合高分辨率的Google Earth影像和GPS地面站点观测数据以及NDVI数据,验证高山林线提取结果的准确性;基于高程数据系统分析研究区高山林线分布与地形特征的关系。结果表明：（1）本文结果与Google Earth影像中实际林线分布基本一致,进一步说明本文提出的林线搜索算法的优越性;（2）秦岭林线高程分布特征呈现显著的坡向差异,呈现南坡林线高于北坡,东坡林线高于西坡的特征。鉴于遥感技术的大范围对地观测的能力,以及卫星影像数据较高的数据质量和易获取性,本文提出的分块迭代搜索策略适用于寻找不同级别的高山林线分布,可进一步推广至全球高山林线制图研究,以期为全球山地生态系统监测、保护和恢复提供技术支撑。     

Assessment of SPOT 5 and QuickBird remotely sensed imagery for mapping tree cover in savannas

The relative abundance and distribution of trees in savannas has important implications for ecosystem function. High spatial resolution satellite sensors, including QuickBird and IKONOS, have been successfully used to map tree cover patterns in savannas. SPOT 5, with a 2.5 m panchromatic band and 10 m multispectral bands, represents a relatively coarse resolution sensor within this context, but has the advantage of being relatively inexpensive and more widely available. This study evaluates the performance of NDVI threshold and object based image analysis techniques for mapping tree canopies from QuickBird and SPOT 5 imagery in two savanna systems in southern Africa. High thematic mapping accuracies were obtained with the QuickBird imagery, independent of mapping technique. Geometric properties of the mapping indicated that the NDVI threshold produced smaller patch sizes, but that overall patch size distributions were similar. Tree canopy mapping using SPOT 5 imagery and an NDVI threshold approach performed poorly, however acceptable thematic accuracies were obtained from the object based image analysis. Although patch sizes were generally larger than those mapped from the QuickBird image data, patch size distributions mapped with object based image analysis of SPOT 5 have a similar form to the QuickBird mapping. This indicates that SPOT 5 imagery is suitable for regional studies of tree canopy cover patterns.     

Predicting soil organic carbon stocks under commercial forest plantations in KwaZulu-Natal province,South Africa using remotely sensed data

ABSTRACT

Commercial forest plantations are increasing globally, absorbing a large amount of carbon valuable for climate change mitigation. Whereas most carbon assimilation studies have mainly focused on natural forests, understanding the spatial distribution of carbon in commercial forests is central to determining their role in the global carbon cycle. Forest soils are the largest carbon reservoir; hence soils under commercial forests could store a significant amount of carbon. However, the variability of soil organic carbon (SOC) within forest landscapes is still poorly understood. Due to limitations encountered in traditional systems of SOC determination, especially at large spatial extents, remote sensing approaches have recently emerged as a suitable option in mapping soil characteristics. Therefore, this study aimed at predicting soil organic carbon (SOC) stocks in commercial forests using Landsat 8 data. Eighty-one soil samples were processed for SOC concentration and fifteen Landsat 8 derived variables, including vegetation indices and bands were used as predictors to SOC variability. The random forest (RF) was adopted for variable selection and regression method for SOC prediction. Variable selection was done using RF backward elimination to derive three best subset predictors and improve prediction accuracy. These variables were then used to build the RF final model for SOC prediction. The RF model yielded good accuracies with root mean square error of prediction (RMSE) of 0.704 t/ha (16.50% of measured mean SOC) and 10-fold cross-validation of 0.729 t/ha (17.09% of measured mean SOC). The results demonstrate the effectiveness of Landsat 8 bands and derived vegetation indices and RF algorithm in predicting SOC stocks in commercial forests. This study provides an effective framework for local, national or global carbon accounting as well as helps forest managers constantly evaluate the status of SOC in commercial forest compartments.     

Soil salinity and vegetation cover change detection from multi-temporal remotely sensed imagery in Al Hassa Oasis in Saudi Arabia

Detecting soil salinity changes and its impact on vegetation cover are necessary to understand the relationships between these changes in vegetation cover. This study aims to determine the changes in soil salinity and vegetation cover in Al Hassa Oasis over the past 28 years and investigates whether the salinity change causing the change in vegetation cover. Landsat time series data of years 1985, 2000 and 2013 were used to generate Normalized Difference Vegetation Index (NDVI) and Soil Salinity Index (SI) images, which were then used in image differencing to identify vegetation and salinity change/no-change for two periods. Soil salinity during 2000–2013 exhibits much higher increase compared to 1985–2000, while the vegetation cover declined to 6.31% for the same period. Additionally, highly significant (p < 0.0001) negative relationships found between the NDVI and SI differencing images, confirmed the potential long-term linkage between the changes in soil salinity and vegetation cover.     

Change detection of linear features in temporally spaced remotely sensed images using edge-based grid analysis

Automatic change detection of land cover features using high-resolution satellite images, is a challenging problem in the field of intelligent remote sensing data interpretation, and is becoming more and more effective for its applications viz. urban planning and monitoring, disaster assessment etc. In the present study, a change in detection approach based on the image morphology that analyses change in the local image grids is proposed. In this approach, edges from both the images are extracted and grid wise comparison is made by probabilistic thresholding and power spectral density analysis for identifying change area. One of the advantages of the proposed methodology is that the temporal images used in the change analysis need not be radiometrically corrected as analysis is based on edge extractions. The grid-based analysis further reduces the error, which might have been introduced by image mis-registration. The proposed methodology is validated by finding the temporal changes in the linear land cover features in parts of Kolkata city, India using three different image data-sets from LISS IV, Cartosat-1 and Google earth having varied spatial resolutions of 5.8 m, 2.5 m and about 1 m, respectively. The overall accuracy in identifying changes is found to be 64.82, 73.86 and 80.93% for LISS IV, Cartosat-1 and Google earth data-set, respectively.     

遥感反演蒸散发的日尺度扩展方法研究进展

遥感技术能够提供卫星过境时刻地表参量的瞬时值,进而通过模型构建可反演得到瞬时蒸散发。相对于瞬时蒸散发,日尺度蒸散发在实际生产生活中具有更重要的应用价值。本文系统地总结分析了遥感反演瞬时蒸散发的代表性日尺度扩展方法,包括蒸发比不变法、解耦因子不变法、辐射能量比不变法、参考蒸发比不变法、地表阻抗不变法和数据同化法,并对各方法的基本原理、估算精度、适用性等进行了对比分析。在此基础上,进一步综述了日尺度扩展方法存在的不确定性和主要问题,包括扩展方法本身误差、云覆盖、气象数据获取、夜间蒸散发估算、遥感反演同扩展误差累积及真实性检验等,并指出今后应从加强有云天及夜间蒸散发扩展机理和方法等方面的研究来提升瞬时蒸散发日尺度扩展精度。     

基于多尺度融合特征卷积神经网络的遥感图像飞机目标检测

飞机检测在遥感图像解译中具有重要的研究意义。针对现有目标检测算法对于复杂场景区域或飞机密集区域的小尺度飞机目标检测精度较低的问题,本文提出了一种端到端的多尺度特征融合飞机目标检测框架MultDet。该方法基于SSD多尺度检测框架,采用轻量级基础网络提取多尺度特征信息;然后设计反卷积特征融合模块,通过跳跃连接将高层语义特征与低层细节特征进行特征融合,得到结构层次丰富的多尺度融合特征;最后设计了一系列不同纵横比的候选框以适应多尺度飞机目标检测。本文在光学遥感图像数据集UCAS-AOD上进行数据分析试验,结果表明,MultDet512在飞机数据集上取得了94.8%的平均检测精度(average precision,AP),在Titan Xp GPU上达到0.0500s/img的检测速度。本文所提飞机目标检测算法在包含多种复杂场景的遥感图像中,能够实现多尺度飞机目标的高精度稳健检测。     

Fuzzy mapping of tropical land cover along an environmental gradient from remotely sensed data with an artificial neural network

Remote sensing is the only feasible means of mapping and monitoring land cover at regional to global scales. Unfortunately the maps are generally derived through the use of a conventional 'hard' classification algorithm and depict classes separated by sharp boundaries. Such approaches and representations are often inappropriate particularly when the land cover being represented may be considered to be fuzzy. The definition of boundaries between classes can therefore be difficult from remotely sensed data, particularly for continuous land cover classes which are separated by a fuzzy boundary which may also vary spatially in time. In this paper a neural network was used to derive fuzzy classifications of land cover along a transect crossing the transition from moist semi-deciduous forest to savanna in West Africa in February and December 1990. The fuzzy classifications revealed both sharp and gradual boundaries between classes located along the transect. In particular, the fuzzy classifications enabled the definition of important boundary properties, such as width and temporal displacement.     

A coupled atmospheric and topographic correction algorithm for remotely sensed satellite imagery over mountainous terrain

Radiometric correction is an important issue in the quantitative remote-sensing community. By integrating dark object subtraction (DOS)-based atmospheric correction with physics-based topographic correction, a coupled land surface reflectance retrieval algorithm (coupled atmospheric and topographic correction algorithm, named the CAT algorithm) for rugged mountainous regions is proposed. Terra MODIS-derived atmospheric characterization data (including aerosol optical depth, integrated precipitable water, surface pressure, and ozone concentration) are employed as inputs for the proposed algorithm. A physics-based path radiance estimation model is proposed and embedded in the CAT algorithm, and band-specific per-pixel path radiance values are calculated. After the CAT algorithm was performed, the correlation between reflectance and terrain was dramatically reduced, with correlation coefficients nearly equal zero, especially for the near infrared and short-wave infrared bands, meanwhile the image information content increased over 20%. To provide a comparison with previous studies, two commonly used methods in the literature (DOS + Cosine and DOS + C) were employed. The results of the comparison show that the proposed algorithm performed better in both atmospheric and topographic corrections without empirical regression.     

Predicting the spatial distribution of hunted and non-hunted Sable antelope (Hippotragus niger niger) using remotely sensed woody cover in a Southern African savanna

From remotely sensed woody cover, we tested whether sables under hunting pressure preferred closed woodland habitats and whether those not under hunting preferred more open woodland habitats. We applied a two factorial logistic regression analysis to model the probability of occurrence of sable antelope in hunted and non-hunted areas of northwest Zimbabwe as a function of vegetation cover density (estimated by a normalized difference vegetation index (NDVI)). We validated the results by high-spatial resolution imagery derived tree canopy area. We subsequently compared the predictions from the two models in order to compare sable cover selection between hunted and non-hunted areas. Our results suggest that hunted sables are likely to select closed woodland, while non-hunted ones would prefer more open woodland habitats. We also established a significant positive relationship between NDVI and tree canopy cover, thus emphasizing the importance of remote sensing in studies that measure the impact of hunting on habitat selection of targeted species.     

Quantifying spatiotemporal dynamics of agricultural landscapes using remotely sensed data and landscape metrics

This study focuses on the spatiotemporal dynamics of agricultural lands and differences in rapidly developing urban and declining rural counties in Iowa, USA between 1984 and 2000. The study presents an analysis of land-cover maps derived from Landsat TM and ETM+ satellite imagery and different landscape metrics using FRAGSTATS and IDRISI software. The study provides evidence of both loss of croplands and change in fragmentation between 1984 and 2000. Fragmentation in agriculture-dominated areas increased with the development of urban centres and diversification of land uses. Fragmentation of landscapes, including agricultural land, was found to be higher in the urbanized counties, but was stable or even declined over time in these counties. In contrast, in the context of remote rural areas, agricultural landscapes experienced rapid increase in fragmentation and farmland loss. The urban–rural gradient analysis used in this study showed that the highest fragmentation occurred on the city edges. These findings suggest that farmland fragmentation is a complex process associated with socio-economic trends at regional and local scales. In addition, socio-economic determinants of landscape fragmentation differ between areas with diverging development trajectories. Intensive cropland fragmentation in remote agricultural regions, detected by this research, should be further studied and its possible effects on both agricultural productivity and biodiversity should be carefully considered.     

面向对象的多尺度加权联合稀疏表示的高空间分辨率遥感影像分类

针对高空间分辨率遥感影像中的地物具有多尺度特性,以及各个尺度的对象特征对地物分类精度的影响具有较强的尺度效性,并结合面向对象影像分析方法和多尺度联合稀疏表示方法在高空间分辨率遥感影像分类中的各自优点,提出了一种面向对象的多尺度加权稀疏表示的高空间分辨率遥感影像分类算法。首先,采用多尺度分割算法获得多尺度分割结果并提取对象的多尺度特征;然后,根据影像对象的多尺度分割质量测度计算各尺度的对象权重,构建面向对象的多尺度加权联合稀疏表示模型;最后,采用2个国产GF-2高空间分辨率遥感数据集和1个高光谱-高空间分辨率航空遥感数据集（WashingtonD.C.数据）验证该算法的有效性。试验结果表明,与SVM、像素级稀疏表示、单尺度和多尺度对象级稀疏表示和深度学习等算法相比较,本文算法获得了较高的OA和Kappa分类精度,提高了各个尺度地物的分类精度,有效抑止了地物分类结果中的椒盐噪声现象,同时保持大尺度地物的区域性和小尺度地物的细节信息。     

小波变换在遥感影像道路特征提取中的应用

遥感影像中的信息提取一直是图像处理的研究热点之一,阐述了在道路特征提取方面的一些经验.鉴于小波变换是检测边缘的良好工具,并结合边界跟踪,最终成功完成了道路特征的提取.     

一种GIS空间知识分析的影像纹理尺度提取算法

本文提出了一种基于GIS领域空间知识的影像纹理尺度提取方法:该方法分析了GIS领域知识辅助下人类对纹理特征的认知过程,对4个步骤的具体实现算法进行探讨;并选取海南省昌江县作为研究区域,实验证明了本方法的有效性,且与传统的枚举法相比,在性能相近的情况下其效率显著提高,更适合于大数据量遥感影像分类运算。     

多尺度SLIC-GMRF与FCNSVM联合的高分影像建筑物提取

遥感影像建筑物提取具有重要的应用价值。然而,高分辨率遥感影像中细节信息繁多、特征复杂,增加了建筑物提取难度。针对这一问题,本文提出一种基于多尺度SLIC-GMRF和FCNSVM的建筑物提取方法,一定程度上提高了高分辨率遥感影像建筑物提取能力。首先,利用多尺度SLIC-GMRF分割算法确定初始建筑物区域,然后,充分利用FCN神经网络在语义分割中的优势抽取建筑物特征,最后,结合提取出的建筑物特征训练SVM分类器细化建筑物提取结果,通过3种控制实验,两种对比方法得出以下结论:SLIC分割算法影响初始分割结果;SVM分类器影响建筑物细部提取;FCN特征影响SVM分类器性能。对于特征清晰、遮挡干扰较少的研究区,本文方法能够较好提取影像中的建筑物,查准率、查全率、质量指标均优于对比方法,对建筑物复杂分布的研究区同样能够取得较好的提取效果。     

高光谱遥感影像多级联森林深度网络分类算法

高光谱遥感技术在环境监测、应急保障、精细地物提取等方面有着广泛的应用,随着高分五号高光谱数据的正式发布,高光谱遥感技术将发挥更重要的作用。遥感影像分类作为高光谱遥感影像信息处理的重要部分,已成为当前研究重点。本文针对传统多级联森林深度学习中模型复杂、无法利用基分类器差异信息、对类间差异较小的样本无法正确区分等不足,提出了一种改进的多级联森林深度学习模型,在模型框架中,分别采用了随机森林和旋转森林作为基分类器,并引入逻辑回归分类器作为判别器用于训练层扩展。相较于传统的深度神经网络,改进的多级联森林深度网络超参数较少且能够自适应确定训练层,更方便进行模型优化。实验采用了高分五号数据集及两个公开的高光谱数据集(Indian Pines数据集及Pavia University数据集)进行精度评定,同时选择了传统分类器支持向量机、深度置信网等模型作为对比分析。实验结果表明,改进的多级联森林深度学习模型能有效地进行高光谱遥感影像分类,且较传统的分类方法精度有所提升。     

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.     

Thematic mapping from remotely sensed data with neural networks: MLP, RBF and PNN based approaches

 Neural networks are attractive tools for the derivation of thematic maps from remotely sensed data. Most attention has focused on the multilayer perceptron (MLP) network but other network types are available and have different properties that may sometimes be more appropriate for some applications. Here a MLP, radial basis function (RBF) and probabilistic neural network (PNN) were used to classify remotely sensed data of an agricultural site. The accuracy of these classifications ranged from 86.25–91.25%. The accuracy of the PNN classification could be increased through the incorporation of prior probabilities of class membership but the accuracy of each classification could also be degraded by the presence of an untrained class. Post-classification analyses, however, could be used to identify potentially misclassified cases, including those belonging to an untrained class, to increase accuracy. The effect of the post-classification analysis on the accuracy of the classification derived from each of the three network types investigated differed and it is suggested that network type be selected carefully to meet the requirements of the application in-hand. Received: 23 March 2000 / Accepted: 9 July 2000