基于全局敏感性分析与机器学习的冬小麦叶面积指数估算

在小麦叶面积指数(leaf area index, LAI)的估算过程中,光谱变量与机器学习算法(MLs)相结合的方法具有较好的性能,但由于输入参数过多会导致数据冗余,使得计算效率降低。为了提高LAI估算的精度和MLs的计算效率,本研究提出了全局敏感性分析(global sensitivity analysis, GSA)与MLs相结合的方法(简称GSA-MLs)。首先,基于PROSAIL模拟数据集,利用GSA量化植被生长参数对Sentinel-2光谱变量的影响;此外利用4种变量筛选策略对所有光谱变量进行排序,并选择最优变量作为MLs的输入参数。然后,通过偏最小二乘回归(partial least square regression, PLSR)、支持向量机(support vector machine, SVM)和随机森林(random forest, RF)3种MLs对小麦叶面积指数(LAI)进行估算。结果表明：红边植被指数主要受叶绿素含量的影响,而短波红外相关的植被指数主要受等效水厚度的影响,所有光谱变量均会受到参数之间的交互作用。SLAI-SInteraction筛选得到的30...     

Leaf Area Index evaluation in vineyards using 3D point clouds from UAV imagery

Precision Agriculture - The Leaf Area Index (LAI) is an ecophysiology key parameter characterising the canopy-atmosphere interface where most of the energy fluxes are exchanged. However, producing...     

Leaf area index based nitrogen diagnosis in irrigated lowland rice

Leaf area index(LAI)is used for crop growth monitoring in agronomic research,and is promising to diagnose the nitrogen(N)status of crops.This study was conducted to develop appropriate LAI-based N diagnostic models in irrigated lowland rice.Four field experiments were carried out in Jiangsu Province of East China from 2009 to 2014.Different N application rates and plant densities were used to generate contrasting conditions of N availability or population densities in rice.LAI was determined by LI-3000,and estimated indirectly by LAI-2000 during vegetative growth period.Group and individual plant characters(e.g.,tiller number(TN)and plant height(H))were investigated simultaneously.Two N indicators of plant N accumulation(NA)and N nutrition index(NNI)were measured as well.A calibration equation(LAI=1.7787LAI_(2000)–0.8816,R~2=0.870~(**))was developed for LAI-2000.The linear regression analysis showed a significant relationship between NA and actual LAI(R~2=0.863~(**)).For the NNI,the relative LAI(R~2=0.808~(**))was a relatively unbiased variable in the regression than the LAI(R~2=0.33~(**)).The results were used to formulate two LAI-based N diagnostic models for irrigated lowland rice(NA=29.778LAI–5.9397;NNI=0.7705RLAI+0.2764).Finally,a simple LAI deterministic model was developed to estimate the actual LAI using the characters of TN and H(LAI=–0.3375(TH×H×0.01)~2+3.665(TH×H×0.01)–1.8249,R~2=0.875~(**)).With these models,the N status of rice can be diagnosed conveniently in the field.     

Relevance of sink-size estimation for within-field zone delineation in vineyards

Source to sink size ratio, i.e.: the relative abundance of photosynthetically active organs (leaves) with regards to photosynthate demanding organs (mainly bunches), is widely known to be one of the main drivers of grape oenological quality. However, due to the difficulty of remote sink size estimation, Precision Viticulture (PV) has been mainly based on within-field zone delineation using vegetation indices. This approach has given only moderately satisfactory results for discriminating zones with differential quality. The aim of this work was to investigate an approach to delineate within-vineyard quality zones that includes an estimator of sink size in the data-set. The study was carried out during two consecutive seasons on a 4.2 ha gobelet-trained cv. ‘Tempranillo’ vineyard. Zone delineation was performed using Normalized Difference Vegetation Index (NDVI), soil apparent electrical conductivity (EC_a) and bunch number (BN) data. These variables were considered separately, in pairs, or the three altogether, using fuzzy k-means cluster analysis for combinations. The zones delineated based on single variables did not allow a sufficient discrimination for grape composition at harvest, NDVI being the only variable that by itself resulted in zones that to some extent differed in grape composition. On the contrary, when two variables were combined, discrimination in terms of grape composition improved remarkably, provided the sink size estimation variable (BN) was included in the combination. Lastly, the combination of the three variables yielded the best discriminating zoning, improving slightly on those provided by NDVI + BN and EC_a + BN combinations. Thus, the relevance of including a variable related to sink size (in this case the number of bunches per plant) has been confirmed, which makes its consideration highly advisable for any PV work aiming at zone delineation for grape quality purposes.     

基于无人机多光谱遥感的冬小麦叶面积指数反演

以获取的冬小麦无人机多光谱影像为数据源,充分利用多光谱传感器的红边通道对传统植被指数进行改进,通过灰色关联度分析后基于多个植被指数建模的方法对冬小麦的叶面积指数(leaf area index,LAI)进行反演精度对比。结果显示:使用基于多植被指数的随机森林(RF)比赤池信息量准则-偏最小二乘法(AIC-PLS)反演精度高。得到的LAI反演值和真实值之间的R~2=0.822,RMSE=1.218。研究证明通过随机森林预测具有更好的拟合效果,对冬小麦的LAI反演有较好的适用性。     

GF-1/WFV在玉米叶面积指数估算中的应用研究

叶面积指数(leaf area index,LAI)是植被冠层重要的结构参数之一,与冠层生理过程密切相关,也是植被遥感领域关注的重要参数之一.本研究对已在轨运行7年的高分一号卫星WFV传感器的植被监测性能进行评测,以吉林省农安县典型玉米分布区作为研究区域,结合地面同步观测的叶面积指数和冠层光谱等实测数据,借助归一化植被...     

Estimating the crop leaf area index using hyperspectral remote sensing

The leaf area index(LAI) is an important vegetation parameter,which is used widely in many applications.Remote sensing techniques are known to be effective but inexpensive methods for estimating the LAI of crop canopies.During the last two decades,hyperspectral remote sensing has been employed increasingly for crop LAI estimation,which requires unique technical procedures compared with conventional multispectral data,such as denoising and dimension reduction.Thus,we provide a comprehensive and intensive overview of crop LAI estimation based on hyperspectral remote sensing techniques.First,we compare hyperspectral data and multispectral data by highlighting their potential and limitations in LAI estimation.Second,we categorize the approaches used for crop LAI estimation based on hyperspectral data into three types:approaches based on statistical models,physical models(i.e.,canopy reflectance models),and hybrid inversions.We summarize and evaluate the theoretical basis and different methods employed by these approaches(e.g.,the characteristic parameters of LAI,regression methods for constructing statistical predictive models,commonly applied physical models,and inversion strategies for physical models).Thus,numerous models and inversion strategies are organized in a clear conceptual framework.Moreover,we highlight the technical difficulties that may hinder crop LAI estimation,such as the curse of dimensionality and the ill-posed problem.Finally,we discuss the prospects for future research based on the previous studies described in this review.     

A simple index to determine if within-field spatial production variation exhibits potential management effects: application in vineyards using yield monitor data

Precision Agriculture - As agricultural data and information becomes more abundant, diagnostics are needed to quickly and efficiently interrogate these data. Indices exist to identify sensor data...     

基于面向对象的橡胶分布面积估算研究

海南是最大的天然橡胶种植区域,常规的调查方法费时费力。随着高分辨率遥感影像的应用,为橡胶信息提取提供了可能。以高分辨率QuickBird卫星影像为基础,采用面向对象的信息提取方法,用光谱、形状、纹理等构建特征空间,进行橡胶的分类试验,获取橡胶的种植面积信息,结果显示,基于面向对象方法的提取精度高,用面向对象的分类方法进行橡胶信息的提取是可行的。     

Evaluation of the use of LIDAR laser scanner to map pruning wood in vineyards and its potential for management zones delineation

Vine vigour assessment has been a major concern of precision viticulture studies in order to identify areas of uniform vine performance within vineyards. Moreover, the counting and weighing of winter dormant canes is considered as the most informative measurement to indicate vine balance and is commonly performed manually by grape growers for management purposes. The main concern of this measurement is that it is time consuming and laborious and it cannot accommodate detailed sampling density. In the present study, the potential of using laser scanner technology as an automated, easy and rapid way to perform mapping of the winter pruning wood across the vineyard was investigated. The study was conducted during 2010 and 2011, in a one hectare commercial vineyard in central Greece, planted with cv. Agiorgitiko, a traditional Greek variety for the production of red wine. Parameters of topography, soil depth, soil texture, canopy properties (NDVI), yield, and grape quality were mapped and analysed in conjunction to winter canes weighing at pruning time. The mapping of the dormant canes was carried out using a 2D laser scanner sensor prior to pruning and manually measuring the pruning weight on a 10 × 20 m grid. Laser scanner measurements showed significant relationship in both 2010 and 2011 with pruning weight (r = 0.809 and r = 0.829 respectively, p < 0.001), yield and early season NDVI, showing the potential of using laser scanner measurements to assess variability in vine vigour within vineyards. These results suggest that laser scanners offer great promise to characterize within field variability in vine performance.     

应用遥感数据反演针叶林有效叶面积指数研究

以红壤丘陵典型区千烟洲及其周边为研究区,利用陆地卫星TM图像数据和同期野外实测的37个针叶林样地有效叶面积指数数据,分析了遥感植被指数与湿地松、杉木林、马尾松和针叶林总体之间的相关关系,进而分别建立了遥感植被指数与不同林型针叶林有效叶面积指数间的线性与非线性回归模型.研究表明,遥感植被指数与不同林型针叶林有效叶面积指数存在较好的相关性,但不同林型之间的相关系数存在一定差异;所建立的针叶林有效叶面积指数遥感反演模型以三次曲线回归方程拟合精度最高.     

Estimation of surface soil organic matter using a ground-based active sensor and aerial imagery

Active canopy sensors are currently being studied as a tool to assess crop N status and direct in-season N applications. The objective of this study was to use a variety of strategies to evaluate the capability of an active sensor and a wide-band aerial image to estimate surface soil organic matter (OM). Grid soil samples, active sensor reflectance and bare soil aerial images were obtained from six fields in central Nebraska before the 2007 and 2008 growing seasons. Six different strategies to predict OM were developed and tested by dividing samples randomly into calibration and validation datasets. Strategies included uniform, interpolation, universal, field-specific, intercept-adjusted and multiple-layer prediction models. By adjusting regression intercept values for each field, OM was predicted using a single sensor or image data layer. Across all fields, the uniform and universal prediction models resulted in less accurate predictions of OM than any of the other methods tested. The most accurate predictions of OM were obtained using interpolation, field-specific and intercept-adjusted strategies. Increased accuracy in mapping soil OM using an active sensor or aerial image may be achieved by acquiring the data when there is minimal surface residue or where it has been excluded from the sensor’s field-of-view. Alternatively, accuracy could be increased by accounting for soil moisture content with supplementary sensors at the time of data collection, by focusing on the relationship between soil reflectance and soil OM content in the 0–1 cm soil depth or through the use of a subsurface active optical sensor.     

Phenotyping a diversity panel of quinoa using UAV-retrieved leaf area index,SPAD-based chlorophyll and a random forest approach

Given its high nutritional value and capacity to grow in harsh environments, quinoa has significant potential to address a range of food security concerns. Monitoring the development of phenotypic traits during field trials can provide insights into the varieties best suited to specific environmental conditions and management strategies. Unmanned aerial vehicles (UAVs) provide a promising means for phenotyping and offer the potential for new insights into relative plant performance. During a field trial exploring 141 quinoa accessions, a UAV-based multispectral camera was deployed to retrieve leaf area index (LAI) and SPAD-based chlorophyll across 378 control and 378 saline-irrigated plots using a random forest regression approach based on both individual spectral bands and 25 different vegetation indices (VIs) derived from the multispectral imagery. Results show that most VIs had stronger correlation with the LAI and SPAD-based chlorophyll measurements than individual bands. VIs including the red-edge band had high importance in SPAD-based chlorophyll predictions, while VIs including the near infrared band (but not the red-edge band) improved LAI prediction models. When applied to individual treatments (i.e. control or saline), the models trained using all data (i.e. both control and saline data) achieved high mapping accuracies for LAI (R²?=?0.977–0.980, RMSE?=?0.119–0.167) and SPAD-based chlorophyll (R²?=?0.983–0.986, RMSE?=?2.535–2.861). Overall, the study demonstrated that UAV-based remote sensing is not only useful for retrieving important phenotypic traits of quinoa, but that machine learning models trained on all available measurements can provide robust predictions for abiotic stress experiments.

     

应用遥感数据反演针叶林有效叶面积指数

以红壤丘陵典型区千烟洲及其周边为研究区,利用陆地卫星TM图像数据和同期野外实测的37个针叶林样地有效叶面积指数数据,分析了遥感植被指数与湿地松、杉木林、马尾松和针叶林总体之间的相关关系,进而分别建立了遥感植被指数与不同林型针叶林有效叶面积指数间的线性与非线性回归模型.研究表明,遥感植被指数与不同林型针叶林有效叶面积指数存在较好的相关性,但不同林型之间的相关系数存在一定差异;所建立的针叶林有效叶面积指数遥感反演模型以三次曲线回归方程拟合精度最高.     

基于无人机激光雷达的人工林碳储量线性与非线性估测模型比较

【目的】森林碳储量是生态系统结构与功能的重要指标,掌握森林碳储量现状有利于森林资源管理。激光雷达能够用于监测森林资源,但是存在森林参数估测的模型多、变量不确定和缺乏林分三维结构解析意义的变量等问题,因此,需要选择合适的林分解析变量和模型。【方法】借助无人机激光雷达点云数据与样地调查数据,以内蒙古自治区赤峰市喀喇沁旗旺业甸人工林为研究对象,分别使用多元线性模型与多元乘幂模型以不同变量对林分碳储量进行估测,选出最优模型并进行精度评价。【结果】研究表明:(1)模型方法而言,非线性模型的检验效果优于线性模型的检验效果:非线性模型(R2为0.66～0.86,rRMSE为23.51%～9.91%),线性模型(R2为0.52～0.85,rRMSE为27.70%～12.38%)。(2)模型使用平均高、郁闭度为基础变量,以穷举法筛选出来的变量组合,估算森林参数得出最佳模型,其中非线性模型以激光点云平均高、郁闭度、高度变动系数和叶面积变动系数的估算精度最高(R2=0.86,rRMSE=9.91%)。【结论】通过激光雷达估测人工林碳储量时,加入垂直结构变量可以提高模型拟合效果,非线性模型比线性模型更适合人工...     

基于机载激光雷达的寒温带典型森林高度制图研究

以内蒙古根河市潮查林场境内的寒温带兴安落叶松原始林及其次生林为研究对象,利用机载激光雷达点云数据与地面调查的66个样地数据,采用不同算法计算样地实测树高(Lorey's高、冠幅面积加权树高和算术平均高)分别与基于双正切角树冠识别算法获取的LiDAR估测高(冠幅面积加权树高、算术平均高)和基于点云提取的百分位高构建树高回归模型(冠幅面积加权树高模型、算术平均树高模型和LiDAR百分位树高模型)。对比不同树高模型的训练精度与估测精度的差异,探讨双正切角树冠识别算法对本研究区的适用性;同时了解冠幅面积加权的样地实测树高与Lorey's高对林分平均高代表性的差异,确定最优解释变量,筛选最优树高模型,计算研究区森林高度空间分布图,为后续生物量和碳储量研究提供参考数据。结果表明:样地冠幅面积加权树高的模型训练精度和估测精度与Lorey's高的结果一致性较好,略低于Lorey's高的估测结果。LiDAR百分位树高模型中的50%分位高与样地实测树高相关性显著且回归模型拟合效果较好,其中,以Lorey's高为样地实测树高时模型的R2=0.869、RMSE=1.366m;以冠幅面积加权树高为样地实测树高时模型的R2=0.839、RMSE=1.392m;Lorey's高的50%分位高模型的估测精度最高,各独立验证样本点估测精度均高于85%,平均估测精度为94.73%,最高估测精度可达99.78%,其中混交林平均估测精度(96.72%)高于针叶林的平均估测精度(93.58%)。因此,选择Lorey's高的50%分位高模型计算研究区的森林高度空间分布。      

An optimized non-linear vegetation index for estimating leaf area index in winter wheat

Precision Agriculture - Using hyperspectral remote sensing technology to monitor leaf area index (LAI) in a timely, fast and non-destructive manner is essential for accurate quantitative crop...     

毛竹林叶面积指数和郁闭度空间分布协同克里格估算

在地面调查的基础上,利用协同克里格插值法对研究区内毛竹Phyllostachys edulis林叶面积指数(LAI,leaf area index)和冠层郁闭度(CC,canopy closure)2个冠层参数进行空间分布估算研究,并与普通克里格插值法进行了比较。研究结果表明:①球状模型可以用来反映LAI和CC的空间变异,且两者具有强烈的空间自相关特征。②协同克里格插值得到的LAI预测值与实测值之间的决定系数R2为0.635 1,而CC的决定系数R2为0.428 5;与普通克里格法相比,基于协同克里格法的LAI和CC预测精度均得到改善,其中LAI预测精度提高了1.94%,均方根误差减少2.00%,平均标准误差减少0.18%,而CC预测精度提高了4.82%,均方根误差减少1.90%,平均标准误差减少1.30%。③安吉县毛竹林LAI和CC都具有从西南到东北逐渐递减空间分布格局,在一定程度上反映了安吉县不同区域毛竹林经营水平的差异。     

B超监测母羊妊娠研究进展

对B超在母羊早期妊娠诊断、母羊怀胎数判定、母羊妊娠日龄预测、胎儿发育监测和胎儿性别鉴定中的应用研究进行了综述,认为超声断层扫描是目前最迅速、最安全、最直观、最有效的母羊妊娠监测手段;对腹壁探查和直肠探查结果进行的对比和总结认为,配种后50～100d是通过腹壁探查进行母羊妊娠诊断的最佳时期,配种后30～30d是通过直肠探查诊断母羊妊娠的最佳时期。     

基于光谱水分指数的阔叶树种叶片等效水厚度估算

植物叶片的水分含量是评价植物水分状态的一项重要指标。为快速、准确地获取植物的叶片含水量,分别使用传统敏感性分析方法和扩展傅立叶振幅灵敏度分析（EFAST方法）对PROSPECT模型进行了敏感性分析。在此基础上筛选与叶片含水量敏感度高的光谱水分指数,并以阔叶树种为研究对象,根据地面实测叶片等效水厚度（l_EWT）和光谱数据建立了叶片等效水厚度估算模型并验证。结果表明:EFAST方法比传统的敏感性分析方法,可以量化叶片各项生化参数的敏感特征。叶片等效水厚度的敏感波段为900~2 500 nm,但在这个范围内光谱反射率也受叶片结构（N）和干物质含量（C_m）的影响;在选取的8个光谱水分指数中,水分胁迫指数（I_MSI）、归一化差异红外指数（I_NDⅡ）和归一化差异水分指数（I_NDWI1640）可以有效提高l_EWT的敏感度。基于实测数据对植被水分指数建立的叶片等效水厚度估算模型的验证结果显示:I_NDWI1640和I_NDⅡ估算模型的预测精度要优于其他指数。对比分析表明:在选取的光谱水分指数中,I_NDWI1640和I_NDⅡ是叶片尺度上估算水分的最佳指数,而归一化差异水分指数（I_NDWI1240）和简单水分指数（I_SRWI）适合于低值区域的叶片等效水厚度反演。