基于高分辨率与高光谱遥感影像的北亚热带马尾松及次生落叶树种的分类

利用遥感数据开展森林资源树种的分类对森林资源的监测、森林可持续经营及生物多样性研究都有重要意义。该文以江苏南部丘陵地区的北亚热带天然次生林为研究对象, 利用LiCHy (LiDAR、CCD、Hyperspectral)集成传感器同期获取的高分辨率和高光谱数据, 进行冠幅识别和多个层次的树种分类: 首先, 对高分辨率影像进行基于边缘检测的多尺度分割, 提取出单木冠幅; 其次, 对高光谱影像进行特征变量提取, 并对提取出的特征变量利用信息熵原理选取优化特征变量; 然后, 分别利用全部特征变量和经优化的重要特征变量对森林树种及森林类型进行预分类; 最后, 在预分类结果中加入单木冠幅信息对森林树种及森林类型进行重分类, 并分析分类结果的精度。研究表明: 1)利用全部特征变量进行4个典型树种分类时, 总体精度为64.6%, Kappa系数为0.493; 而针对森林类型的分类精度为81.1%, Kappa系数为0.584。2)利用选取的优化特征变量分类精度略低于利用全部特征变量的分类精度, 其中对4个典型树种分类时, 总体精度为62.9%, Kappa系数为0.459; 而针对森林类型的分类精度为77.7%, Kappa系数为0.525。通过集成传感器同期获取的高分辨率和高光谱数据可以有效地进行北亚热带森林的树种分类及森林类型的划分。     

面向对象的优势树种类型信息提取技术

森林植被优势树种类型信息的提取是遥感影像分类中的难点.面向对象分类方法是用高空间分辨率遥感数据实现精确类型信息提取的新方法.本文以2013年Quickbird影像作为基础数据,选择福建省三明市将乐林场为研究区,采用面向对象多尺度分割方法提取耕地、灌草地、未成林造林地、马尾松、杉木和阔叶树等类型信息.分类特征融合植被的光谱、纹理和多种植被指数3类特征信息,建立类层次结构,对不同层次分别用隶属度函数和决策树分类规则,最终完成分类,并与只用纹理与光谱特征相结合的方法进行对比.结果表明:融合纹理、光谱、多种植被指数的面向对象的分类方法提取研究区优势树种类型信息的精度为91.3％,比只用纹理和光谱的方法精度提高了5.7％.     

基于多源遥感数据的植物物种分类与识别: 研究进展与展望

物种分类与识别是生物多样性监测的基础, 明确物种的类别及其分布是解决几乎所有生态学问题的前提。为深入了解基于多源遥感数据的植物物种分类与识别相关研究的发展现状和存在的问题, 本文对2000年以来该领域的研究进行了总结分析, 发现: 当前大多数研究集中在欧洲和北美地区的温带或北方森林以及南非的热带稀树草原; 使用最多的遥感数据是机载高光谱数据, 而激光雷达作为补充数据, 通过单木分割及提供单木的三维垂直结构信息, 显著提高了分类精度; 支持向量机和随机森林作为应用最广的非参数分类算法, 平均分类精度达80%; 随着计算机技术及机器学习领域的不断成熟, 人工神经网络在物种识别领域得以迅速发展。基于此, 本文对目前基于遥感数据的植物物种分类与识别中在分类对象复杂性、多源遥感数据整合、植物物候与纹理特征整合和分类算法技术等方面面临的挑战进行了总结, 并建议通过整合多时相监测数据、高光谱和激光雷达数据、短波红外等特定波谱信息、采用深度学习等方法来提高分类精度。     

集成高分辨率UAV影像与激光雷达点云的亚热带森林林分特征反演

该研究集成高分辨率无人机(UAV)影像和激光雷达(LiDAR)点云数据估算亚热带天然次生林林分基本特征变量。首先, 基于LiDAR点云和反距离加权插值法构建林下高精度数字高程模型(DEM); 然后利用UAV影像对序列构建植被冠层上层三维点云, 并借助DEM进行高度信息归一化, 提取高度和冠层点云密度相关的特征变量; 最后, 构建预测模型并估算Lorey’s高、林分密度、胸高断面积、蓄积量。结果表明: 联合提取的特征变量与Lorey’s高的敏感性最高, 蓄积量次之, 林分密度和胸高断面积最低; 利用UAV灵活快速的手段获取森林冠层信息, 辅以高精度LiDAR数据获取的地形信息, 两者互补实现一种可重复的快速、廉价和灵活的林分特征的反演方式。     

基于Sentinel-2A影像的枸杞种植区域识别

枸杞(Lycium barbarum L.)是西北地区重要的经济作物,靖远县作为甘肃省主要枸杞产区之一,快速准确地获取枸杞种植结构与空间分布对当地农业种植结构调整及区域经济可持续具有重要意义。基于Sentinel-2A影像数据,采用面向对象的分类方法,利用光谱特征与纹理特征构建随机森林分类器,实现枸杞种植信息的提取。结果表明：将光谱特征和纹理特征相结合的随机森林分类方法精度最高,分类总体精度达到88.14%,Kappa系数为0.81,枸杞的用户精度为81.03%;靖远县枸杞种植面积为297.12 km~2,分别呈现出集中连片、零星分布特征,主要集中分布在靖安乡、五合镇、东升镇和北滩镇的种植基地,采用Sentinel-2A可以很好地提取空间上分散种植的枸杞。研究结果可为靖远县特色农作物枸杞的种植结构调整和开发利用提供数据支撑,研究方法可为大面积的枸杞遥感监测提供参考。     

神农架川金丝猴栖息地优势乔木树种遥感识别及其分布特征

针对神农架川金丝猴生境基础研究中乔木树种大范围分布数据难以获取问题,尝试利用多源多时相遥感数据结合专家知识分层次实现树种识别。首先采用冬季Landsat8/OLI数据根据物侯特性分层提取常绿、落叶林的地域范围;进而依据夏季Worldview-2高分遥感影像的实地乔木样本的光谱特征分层次完成常绿树种(巴山冷杉、华山松、青$、刺叶栎)和落叶树种(红桦、日本落叶松、米心水青冈、漆树、锐齿槲栎、椅杨)的识别;并通过实地植被样方及专家知识通过高程数据完成分类结果的修正;最后结合GIS对主要优势树种的地形及地域分布特征进行了空间分析。实验精度表明常绿林中巴山冷杉、华山松、刺叶栎、虫害华山松整体精度较高,落叶林中红桦、漆树等识别精度相对较高,部分树种如椅杨、锐齿槲栎识别精度较低;总体上常绿树种的精度要优于落叶树种。从植物地理学、遥感、GIS三者相结合的角度,将多源、多时相遥感数据与物种物候特性、专家知识进行有效整合,提出了一种乔木树种识别的方法(1)提供了复杂山地环境的主要乔木优势种识别途径,且具有通用性;(2)完成了物种物候特性与遥感数据特性的整合利用,有效降低数据成本费用;(3)配合地面样方及专家知识修正结果,避免了过分依赖光谱特征引起的误判。这将为神农架川金丝猴栖息地保护与恢复提供更精确的数据依据。     

无人机高光谱联合LiDAR估测林分与单木尺度叶绿素含量

叶绿素是表征植被健康状况的重要指标,它的准确估计对森林碳汇评价研究至关重要。本研究通过无人机高光谱数据联合激光雷达点云估计针叶林、阔叶林和针阔混交林林分与单木水平的叶绿素含量,提升叶绿素无损估测精度,全面分析不同尺度叶绿素含量空间分布规律。在无人机高光谱数据与激光雷达点云融合的基础上,结合地面样地实测数据,通过相关性分析筛选与叶绿素含量相关的36个光谱特征变量,采用统计模型多元逐步回归、BP神经网络、萤火虫算法优化的BP神经网络、随机森林和混合数据驱动的机理模型PROSPECT模型构建多个叶绿素估算模型,选取最优模型估算森林叶绿素含量,分析其在林分和单木尺度上水平方向与垂直方向的空间分布规律。结果表明：在统计模型中,随机森林(R²=0.59～0.64,RMSE=3.79～5.83μg·cm^-2)优于多元逐步回归、BP神经网络和萤火虫算法优化的BP神经网络构建的模型;机理模型验证精度最高(R²=0.97,RMSE=3.40μg·cm^-2)。不同林分类型叶绿素的含量存在较大差异,阔叶林叶绿素含量为25....     

基于无人机影像与面向对象-深度学习的滨海湿地植物物种分类

明确滨海湿地植物物种类型及其分布状况是实现滨海湿地精细化生物多样性监测的基础,对于滨海湿地的保护管理与生态可持续发展均具有重要意义。本研究以无人机可见光遥感影像为基础数据源,在定量分析最优分割尺度与最优分类特征组合的基础上,应用面向对象-U-net深度学习方法对闽江河口湿地植物物种类型进行分类,并与K最近邻、决策树、随机森林和贝叶斯分类方法进行精度对比分析,以期为滨海湿地植物物种遥感精细分类与生物多样性保护管理提供方法借鉴与科学参考。研究结果表明,利用面向对象-U-net深度学习方法提取不同滨海湿地植物物种类型的分类精度可达95.67%,总体精度较其他分类方法提高6.67%–13.67%, Kappa系数提高0.12–0.31,且分类整体性好。此外,实现植物物种光谱特征、形状特征、纹理特征与高度特征的最优特征选择对于有效提高湿地植物物种信息分类精度具有重要作用,应用最优分割尺度实现影像分割可提高整体分类效率。     

基于环境星与MODIS时序数据的面向对象森林植被分类

林区地形复杂、植被分布无序,且森林植被光谱信息相近,因而森林二级类型边界的确定成为土地覆盖遥感分类的难点。选择吉林省东部山区为研究区,以环境星影像(HJ-1 CCD)和中等分辨率成像光谱仪(MODIS)时序数据为基础,采用面向对象的分类方法进行森林植被类型的提取。分类特征参数主要选取了HJ-1 CCD的光谱和纹理特征,以及MODIS时序数据的物候特征。研究区总体分类精度为91.5%,Kappa系数为0.88,森林二级类型的分类精度均较高,其中落叶阔叶林的制图精度达到了97.1%。所用的面向对象分类方法与未加入物候特征的面向对象分类方法相比,森林二级类型的分类精度得到大幅度提高。     

基于无人机高光谱影像和深度学习算法的长白山针阔混交林优势树种分类

快速、准确识别树种及其分布格局是森林资源经营管理和生物多样性保护的基础和前提。与传统实地调查的方法相比,近年来飞速发展的近地面遥感技术可以灵活、高效和便捷地获取高分辨率高光谱遥感影像,而如何从包含丰富信息的诸多特征中选择信息量大且冗余度低的特征进行树种自动识别,是当前研究亟待解决的问题。本研究以长白山25 hm~2温带针阔混交样地为主要研究平台,于2019年8月使用无人机搭载的光谱传感器获取面积为6 hm~2的高光谱影像,选择红松、春榆、蒙古栎、水曲柳、大青杨和紫椴6种林冠层树种作为实地标记树种,使用实时载波相位差分技术对所选目标树种进行精确定位,结合2019年样地复查结果对研究区的影像进行目视解译,分别使用卷积神经网络法、最大似然法和马氏距离3种分类方法进行冠层树种的自动分类研究。结果表明：(1)卷积神经网络的树种分类总体精度和Kappa系数(99.85%、0.998)优于最大似然法(89.11%、0.86)和马氏距离法(79.65%、0.75)。(2)在3种分类方法中,单个优势树种分类精度均在卷积神经网络中为最高精度,红松、春榆、蒙古栎、水曲柳、大青杨和紫椴的最高分类精度分别为10...     

Classification of Pinus massoniana and secondary deciduous tree species in northern subtropical region based on high resolution and hyperspectral remotely sensed data

Aims Using remote sensing data for tree species classification plays a key role in forestry resource monitoring, sustainable forest management and biodiversity research.Methods This study used integrated sensor LiCHy (LiDAR, CCD and Hyperspectral) to obtain both the high resolution imagery and the hyperspectral data at the same time for the natural secondary forest in south Jiangsu hilly region. The data were used to identify the crown and to classify tree species at multiple levels. Firstly, tree crowns were selected by segmenting high-resolution imagery at multiple scales based on edge detection; secondly, characteristic variables of hyperspectral images were extracted, then optimization variables were selected based on the theory of information entropy. Tree species and forest types were classified using either all characteristic variables or optimization variables only. Finally, tree species and forest types were reclassified along with the tree crowns information, and the accuracy of classification was discussed. Important findings Based on all available characteristic variables, the overall accuracy for four typical tree species classification was 64.6%, and the Kappa coefficient was 0.493. The overall accuracy for forest types classification was 81.1%, and the Kappa coefficient was 0.584. Based on optimization variables only, the overall accuracy for four typical tree species classification dropped to 62.9%, and the Kappa coefficient was 0.459. The overall accuracy for forest types classification was 77.7%, and the Kappa coefficient was 0.525. Obtaining both high resolution image and hyperspectral data at the same time by integrated sensor can increase overall accuracy in classifying forest types and tree species in northern subtropical forest.     

Quantifying Tropical Dry Forest Type and Succession: Substantial Improvement with LiDAR

Improved technologies are needed to advance our knowledge of the biophysical and human factors influencing tropical dry forests, one of the world's most threatened ecosystems. We evaluated the use of light detection and ranging (LiDAR) data to address two major needs in remote sensing of tropical dry forests, i.e., classification of forest types and delineation of forest successional status. We evaluated LiDAR‐derived measures of three‐dimensional canopy structure and subcanopy topography using classification‐tree techniques to separate different dry forest types and successional stages in the Guánica Biosphere Reserve in Puerto Rico. We compared the LiDAR‐based results with classifications made from commonly used remote sensing data, including Landsat satellite imagery and radar‐based topographic data. The accuracy of the LiDAR‐based forest type classification (including native‐ and exotic‐dominated forest classes) was substantially higher than those from previously available data (kappa = 0.90 and 0.63, respectively). The best result was obtained when combining LiDAR‐derived metrics of canopy structure and topography, and adding Landsat spectral data did not improve the classification. For the second objective, we observed that LiDAR‐derived variables of vegetation structure were better predictors of forest successional status (i.e., mid‐secondary, late‐secondary, and primary forests) than was spectral information from Landsat. Importantly, the key LiDAR predictors identified within each classification‐tree model agreed with previous ecological knowledge of these forests. Our study highlights the value of LiDAR remote sensing for assessing tropical dry forests, reinforcing the potential for this novel technology to advance research and management of tropical forests in general.     

普洱季风常绿阔叶林乔木物种多样性高分辨率遥感估测

季风常绿阔叶林是我国南亚热带典型的地带性植被,也是云南省普洱地区重要森林类型。季风常绿阔叶林乔木物种多样性遥感估测对研究区域尺度生物多样性格局及其规律具有重要作用。根据光谱异质性假说和环境异质性假说,首先使用1m空间分辨率的机载高光谱数据和激光雷达数据提取了光谱多样性特征和垂直结构特征。然后利用基于随机森林算法的递归特征消除方法选择对研究区森林乔木物种多样性指数具有较好解释能力的遥感特征,并对Shannon-Winner物种多样性指数进行建模、制图。研究结果表明：(1)基于机载LiDAR数据提取的垂直结构特征和机载高光谱数据提取的光谱多样性特征均对研究区森林乔木物种多样性具有较好的解释能力,随机森林模型估测结果分别为R²=0.48,RMSE=0.46和R²=0.5,RMSE=0.45;两种数据源融合可以进一步提高遥感数据的森林乔木物种多样性估测精度,随机森林估测模型R²和RMSE分别为0.69和0.37。(2)机载激光雷达数据对研究区针阔混交林乔木物种多样性的估测能力优于机载高光谱数据。(3)机器学习方法有助于从高维遥感...     

基于多光谱影像的森林树种识别及其空间尺度响应

当前,不同空间分辨率卫星影像对森林类型识别结果中普遍存在的尺度效应,而且纹理参量对不同尺度下树种识别精度的影响仍缺乏广泛认知.本研究以中国东北旺业甸林场为研究区,采用观测时相同步、地理坐标匹配的GF-1 PMS、GF-2 PMS、GF-1 WFV,以及Landsat-8 OLI卫星传感器数据组成空间尺度观测序列(1、2、4、8、16、30 m),并结合支持向量机(SVM)模型,探讨了区域内5种优势树种遥感识别结果的尺度变化规律及其纹理特征参数的影响,同时检验了基于尺度上推转换影像的树种识别结果差异.结果表明: 影像空间分辨率对区域树种识别结果具有显著影响,其中,研究区森林树种识别的最佳影像分辨率为4 m,当分辨率降低至30 m时,树种识别结果最差.在1～8 m影像分辨率范围内,增加纹理信息能够显著提高不同优势树种的识别精度,使总分类精度提升了2.0%～3.6%,但纹理信息对16～30 m影像的识别结果没有显著影响.与真实尺度卫星影像相比,基于升尺度转换影像的树种识别结果及其尺度响应特征存在显著差异,表明在面向多个空间尺度的遥感观测和应用研究中,需要采用真实分辨率影像以确保结果的准确性.     

Benefits of hyperspectral remote sensing for tracking plant invasions

Aim We aim to report what hyperspectral remote sensing can offer for invasion ecologists and review recent progress made in plant invasion research using hyperspectral remote sensing. Location United States. Methods We review the utility of hyperspectral remote sensing for detecting, mapping and predicting the spatial spread of invasive species. We cover a range of topics including the trade‐off between spatial and spectral resolutions and classification accuracy, the benefits of using time series to incorporate phenology in mapping species distribution, the potential of biochemical and physiological properties in hyperspectral spectral reflectance for tracking ecosystem changes caused by invasions, and the capacity of hyperspectral data as a valuable input for quantitative models developed for assessing the future spread of invasive species. Results Hyperspectral remote sensing holds great promise for invasion research. Spectral information provided by hyperspectral sensors can detect invaders at the species level across a range of community and ecosystem types. Furthermore, hyperspectral data can be used to assess habitat suitability and model the future spread of invasive species, thus providing timely information for invasion risk analysis. Main conclusions Our review suggests that hyperspectral remote sensing can effectively provide a baseline of invasive species distributions for future monitoring and control efforts. Furthermore, information on the spatial distribution of invasive species can help land managers to make long‐term constructive conservation plans for protecting and maintaining natural ecosystems.     

LiDAR GEDI derived tree canopy height heterogeneity reveals patterns of biodiversity in forest ecosystems

The “Height Variation Hypothesis” is an indirect approach used to estimate forest biodiversity through remote sensing data, stating that greater tree height heterogeneity (HH) measured by CHM LiDAR data indicates higher forest structure complexity and tree species diversity. This approach has traditionally been analyzed using only airborne LiDAR data, which limits its application to the availability of the dedicated flight campaigns. In this study we analyzed the relationship between tree species diversity and HH, calculated with four different heterogeneity indices using two freely available CHMs derived from the new space-borne GEDI LiDAR data. The first, with a spatial resolution of 30 m, was produced through a regression tree machine learning algorithm integrating GEDI LiDAR data and Landsat optical information. The second, with a spatial resolution of 10 m, was created using Sentinel-2 images and a deep learning convolutional neural network. We tested this approach separately in 30 forest plots situated in the northern Italian Alps, in 100 plots in the forested area of Traunstein (Germany) and successively in all the 130 plots through a cross-validation analysis. Forest density information was also included as influencing factor in a multiple regression analysis. Our results show that the GEDI CHMs can be used to assess biodiversity patterns in forest ecosystems through the estimation of the HH that is correlated to the tree species diversity. However, the results also indicate that this method is influenced by different factors including the GEDI CHMs dataset of choice and their related spatial resolution, the heterogeneity indices used to calculate the HH and the forest density. Our finding suggest that GEDI LIDAR data can be a valuable tool in the estimation of forest tree heterogeneity and related tree species diversity in forest ecosystems, which can aid in global biodiversity estimation.     

Improved methods for measuring forest landscape structure: LiDAR complements field-based habitat assessment

Conservation and monitoring of forest biodiversity requires reliable information about forest structure and composition at multiple spatial scales. However, detailed data about forest habitat characteristics across large areas are often incomplete due to difficulties associated with field sampling methods. To overcome this limitation we employed a nationally available light detection and ranging (LiDAR) remote sensing dataset to develop variables describing forest landscape structure across a large environmental gradient in Switzerland. Using a model species indicative of structurally rich mountain forests (hazel grouse Bonasa bonasia), we tested the potential of such variables to predict species occurrence and evaluated the additional benefit of LiDAR data when used in combination with traditional, sample plot-based field variables. We calibrated boosted regression trees (BRT) models for both variable sets separately and in combination, and compared the models’ accuracies. While both field-based and LiDAR models performed well, combining the two data sources improved the accuracy of the species’ habitat model. The variables retained from the two datasets held different types of information: field variables mostly quantified food resources and cover in the field and shrub layer, LiDAR variables characterized heterogeneity of vegetation structure which correlated with field variables describing the understory and ground vegetation. When combined with data on forest vegetation composition from field surveys, LiDAR provides valuable complementary information for encompassing species niches more comprehensively. Thus, LiDAR bridges the gap between precise, locally restricted field-data and coarse digital land cover information by reliably identifying habitat structure and quality across large areas.     

航空航天遥感在物种多样性研究与保护中的应用

人类活动导致全球范围内生物多样性丧失日趋严重。物种多样性是研究最为深入以及最贴近生物多样性管理的层次。物种多样性的研究往往受到多时空尺度生态过程的影响, 传统物种多样性调查方法受到人力物力影响, 局限性大, 物种多样性的研究与管理亟需整合不同来源的数据。遥感技术从传统的光学遥感阶段发展到不同平台、不同维度相结合的多源遥感阶段, 并逐渐进入以高空间分辨率和高光谱为特征、以激光雷达为前沿发展方向的综合遥感阶段。遥感技术因为其监测范围广、能监测人迹罕至地区以及长期可重复等特性, 为研究不同时空尺度的生态学科学问题提供了更新更优的研究手段。本文围绕种群动态、种间关系与群落多样性、功能属性及功能多样性以及生物多样性保护管理等生物多样性研究热点问题, 系统地论述了航空航天遥感技术在物种多样性研究与保护领域的应用, 总结了航空航天遥感技术在研究与物种多样性有关的主要生态学问题中的机遇与挑战。我们认为航空航天遥感技术利用多光谱甚至高光谱与激光技术从空中监测物种多样性, 从不同视角、基于不同光源提供了物种多样性不同侧面的信息, 能够减小地面调查强度, 在大范围和边远地区的物种多样性调查研究中有着至关重要的作用。依据光谱特性的物种判别以及依据激光雷达的三维结构量测将促进生物多样性的研究与管理, 加强遥感学家和生物多样性研究者的沟通交流将有助于促进不同时空尺度的生物多样性与遥感技术的结合。