结合多模态数据的滨海湿地碱蓬叶面积指数无人机高光谱反演
UAV hyperspectral inversion of
Suaeda Salsa leaf area index in coastal wetlands combined with multimodal data- 2023年27卷第6期 页码:1441-1453
纸质出版日期: 2023-06-07
DOI: 10.11834/jrs.20222136
引用
阅读全文PDF
扫 描 看 全 文
纸质出版日期: 2023-06-07
扫 描 看 全 文
引用
阅读全文PDF
叶面积指数(LAI)是表征植被对光合辐射的吸收和拦截以及植被生长状况的重要参数。为准确、快速地提高叶面积指数估算精度,选取黄河三角洲碱蓬滩湿地,以中国土著植物碱蓬为研究对象,获取无人机高光谱遥感影像和测定地面光谱,结合区域土壤因子、植被光谱特征、高光谱影像纹理特征和植被覆盖度构建多模态数据,发展随机森林RF(Random Forest)和粒子群PSO(Particle Swarm Optimization)双优化策略的深度极限学习机DELM(Deep Extreme Learning Machine)算法构建滨海湿地碱蓬叶面积指数反演模型,决定系数(R2)和均方根误差(RMSE)分别为0.9546,0.1341。与基于支持向量机SVM(Support Vector Machine)、BP神经网络BP(Back Propagation Neural Network)、极限学习机ELM(Extreme Learning Machine)、深度极限学习机DELM(Deep Extreme Learning Machine)、粒子群优化的深度极限学习机(PSO-DELM)5种算法构建的碱蓬叶面积指数反演模型精度相比,决定系数R2最高提高了0.2654,均方根误差RMSE最大降低了0.0828。与传统的反演模型SVM相比,RF-PSO-DELM模型具有更好的泛化性,融合多模态数据则可以有效提高反演模型精度。该研究进一步丰富了基于无人机高光谱遥感技术实现盐沼植被精准监测的理论和技术。
Salt marsh vegetation is an important part of the blue carbon ecosystem, which has strong carbon sequestration and carbon storage capacity. The Leaf Area Index (LAI) determines its growth, photosynthetic active radiation absorption ratio, and biomass. LAI is often used to simulate the photosynthesis, respiration, and transpiration of vegetation, thus playing a crucial role in improving the yield of Suaeda salsa. An accurate estimation of Suaeda salsa can provide an important basis for judging the growth status of alkali ponies and thus provide an effective aid for monitoring salt marsh wetlands.
To improve the accuracy of LAI estimation accurately and rapidly, the Yellow River Delta Suaeda salsa shoal wetland was selected, and the indigenous plant Suaeda salsa was used as the research object. A UAV hyperspectral remote sensing image was obtained, and the ground spectrum was measured in combination with regional soil factors, vegetation spectral characteristics, hyperspectral image texture characteristics, and vegetation coverage. Multimodal data, through Random Forest (RF) feature selection for multimodal data, and the RF-PSO-DELM algorithm with a dual optimization strategy was developed to construct an inversion model of the LAI of Suaeda salsa in coastal wetlands.
The coefficient of determination (R2) and the Root Mean Square Error (RMSE) were 0.9546 and 0.1341, respectively. Compared with the inversion model accuracy of Suaeda salsa LAI constructed on the basis of the five algorithms of SVM, BP, ELM, DELM, and PSO-DELM, R2 was increased by 0.2654 at most, and the RMSE was reduced by 0.0828 at most.
Compared with the traditional inversion model (SVM), the RF-PSO-DELM model had better generalization; moreover, the fusion of multimodal data could effectively improve the accuracy of the inversion model. This study further enriched the theory and technology for the accurate monitoring of salt marsh vegetation based on UAV hyperspectral remote sensing technology. Multisource modal data, such as soil factors, texture features, spectral features, and vegetation cover affecting the growth of alkali ponies in coastal wetlands, were comprehensively considered, and the important influencing factors sensitive to the LAI of alkali ponies’ LAI were extracted by the random forest feature preference algorithm, which effectively reduced the complexity of model inversion and greatly improved the accuracy of model prediction.
碱蓬是一种滨海湿地改良土壤的重要植物(
相较于传统植被LAI测量方法,无人机高光谱遥感技术因其快速、重复的信息捕获能力及高空间分辨率的特点,在小地块精确监测中被广泛应用。众多学者围绕无人机高光谱影像提出了植被指数法(
由于无人机高光谱数据一般具有多重共线性,存在数据冗余问题,机器学习算法具有较高的学习和预测能力,能够从不同角度克服变量间共线性的问题(
当前研究多侧重基于单一特征(植被指数、纹理特征等)的植被LAI反演研究,而综合多特征因子对植被LAI的反演研究较少。当变量数目较多时,线性回归等方法受到限制,且常用的机器学习方法:随机森林算法RF(Random Forest)易受环境干扰波动大(
本文以滨海湿地碱蓬为研究对象,采集无人机高光谱遥感影像和地面实测碱蓬反射光谱、叶面积指数及区域土壤因子,发展随机森林(RF)和粒子群(PSO)双优化策略的深度极限学习机(DELM)算法结合多模态数据的算法开展盐沼湿地碱蓬LAI的无人机高光谱反演研究。RF多模态特征优选可解决高光谱数据中存在的冗余特征和可分性较差的问题,有效优选出敏感特征;PSO-DELM快速、准确地对多模态数据进行反演建模,提高反演模型精度,为盐沼湿地碱蓬的实时监测和生态修复提供技术支撑。
选取山东省东营市黄河三角洲自然保护区内一块长为600 m,宽为400 m的矩形碱蓬生长密集区作为研究区(37°47′00″N—37°47′30″N,119°09′30″E—119°10′00″E)(
图1 研究区地理位置及采样点空间分布
Fig. 1 The geographical location of the study area and the spatial distribution of sampling points
经前期实地考察发现区域内地势平坦,碱蓬植被株高为10—20 cm,冠径大小为5—15 cm,研究区南部碱蓬生长比较集中,北部碱蓬比较稀疏。根据碱蓬生长状况,共划分了密度不等的123个采样区,每个采样区为1 m2的样方。为了更好地了解研究区土壤养分含量,用对角线采样法在样方内对0—20 cm水平的表层土壤进行采样并将其均匀混合作为土壤样品,并测定土壤中的pH、全盐、全氮、全磷、全钾、有机质、总碳、含水量。
2.2.1 无人机高光谱图像获取和预处理
无人机遥感平台采用大疆M600Pro六旋翼无人机,无人机搭载的高光谱传感器为北京智科远达数据技术有限公司自主研发的新型ZK-VNIR-FPG480高光谱成像仪(
图2 标准假彩色影像(R: 846nm, G: 641 nm, B:538 nm)
Fig. 2 Standard false color image (R: 846 nm, G: 641 nm, B: 538 nm)
本研究区位于黄河三角洲国家级自然保护区内一块碱蓬生长密集区域,且由
为了提取无人机高光谱影像中每个样方的冠层光谱反射率,先在每个样方中碱蓬覆盖区建立感兴趣区,通过统计分析,将其平均值作为碱蓬样方的高光谱反射率。经统计发现无人机高光谱遥感影像在400—450 nm和900—1000 nm波段范围内的光谱反射率噪声较大,因此选取450—900 nm波段范围内的198个波段进行碱蓬LAI反演研究。
无人机高光谱遥感图像在获取过程中因受环境因素的影响会不可避免地产生噪声,对无人机高光谱遥感影像依次进行S-G平滑滤波、包络线去除和一阶微分处理。首先,利用ENVI软件中的S-G平滑滤波工具对无人机高光谱遥感影像进行平滑滤波,窗口设为60,拟合多项式设为5;之后,将平滑滤波处理后的无人机高光谱影像进行包络线去除,并将每个样方的平均反射光谱作为样方碱蓬包络线去除后的反射光谱;最后,将包络线去除后的每个样方碱蓬平均反射光谱进行一阶微分处理,以突出对碱蓬LAI敏感的特征波段。
2.2.2 特征提取
基于无人机高光谱影像提取纹理特征以及碱蓬反射光谱提取红边参数和植被指数作为碱蓬LAI反演的多模态数据。
(1)红边参数提取。依据
将计算后的红边参数与碱蓬LAI进行相关性分析,结果如
(2)植被指数提取。将碱蓬冠层反射光谱进行任意波段组合分别构建比值植被指数RVI(
图3 预处理后的3种VI与LAI的秩序相关系数
Fig. 3 Order correlation coefficients between three vegetation indexes and LAI after preprocessing
DRVI=DxDy | (1) |
DDVI=Dx-Dy | (2) |
DNDVI=Dx-DyDx+Dy | (3) |
式中,D为一阶微分后的光谱反射率,x,y为450—900 nm的任意波段。
因此,植被指数D690/D598、(D699-D598)和(D598-D693)/(D598+D693)可作为构建碱蓬LAI反演模型的敏感植被指数。
(3)纹理特征提取。计算无人机高光谱影像各波段反射率与LAI之间的皮尔逊相关系数(Pearson Correlation Coefficient),结果如
图4 碱蓬冠层光谱反射率与LAI之间的秩序相关系数
Fig. 4 Order correlation coefficients between reflectance and LAI after preprocessing
利用灰度共生矩阵方法基于对700 nm的无人机高光谱波段提取均值、方差、协同性、对比度、相异性、信息熵、二阶矩和相关性8个纹理特征。综合考虑无人机高光谱影像的空间分辨率(0.1 m)和区域碱蓬分布特点,选取3×3窗口;因无人机航向接近45°,在纹理分析时方向选取45°方向分析8个纹理特征,同时将45°方向的8个纹理特征作为碱蓬LAI反演模型的敏感纹理特征。
2.2.3 碱蓬LAI测定
在获取无人机高光谱遥感图像的同时,同步测定了碱蓬叶面积指数。为了确保仪器的探测镜头视场角完全覆盖样方内所有碱蓬,在每个样方的中心位置挖出一个直径15 cm,深15 cm的小洞,然后把CI-110植物冠层分析仪器探头置于洞内,且不要接触洞的四周,并保证探头顶部与地面平行。使用CI-110冠层分析仪自带数据处理软件(Plant Canopy Analyst System)计算出每个样方的碱蓬LAI。
2.2.4 土壤因子测定
将采集的土壤样品分为两部分,其中一部分原样用于测定土壤含水量;另一部分土壤原样,首先进行自然风干,然后充分混合、过筛和保存备用,之后测定土壤pH、全盐、全氮、全磷、全钾、有机质、总碳等要素(
RF是一种将回归学习方法与Bagging技术相结合的集成分类算法(
由于各种各样的环境因素和无人机高光谱等高维多模态数据都会影响碱蓬LAI,且RF可以高效地处理高维数据的非线性关系,因此本文采用RF筛选出对碱蓬LAI影响程度较高的多模态特征变量。
经多次实验验证后,RF模型具体参数设置如
PSO是一种并行的智能群集优化算法(
极限学习机ELM(Extreme Learning Machine)是隐藏层节点参数随机初始化,最小二乘法求解的单隐含层前馈神经网络,具有学习速度快和泛化能力强等优点(
DELM以自编码训练方式为基础,利用分层的无监督方法提取高维数据的隐藏信息,并以ELM作为基础学习单位,通过多个ELM级联得到以层次编码形式传输的最优网络参数,将定量特征通过ELM映射到输出空间,在维持彼此量化特征的同时,最大限度地减少重构误差。DELM避免反向调优过程,减少数据损失,有效提高模型预测精度。
尽管DELM有助于多层次学习特征,减少了调整参数的需要,然而输入的权重和阈值是正交矩阵随机产生的,从而导致预测过程中模型不稳定,波动较大。因此,采用PSO智能优化算法对初始权重和阈值进行参数寻优。
PSO-DELM模型整个流程步骤说明如下:
(1)初始化粒子群参数。随机初始化粒子的位置和速度,粒子的迭代次数、种群规模、种群维度和惯性因子分别设置为200、40、2和0.9。
(2)以均方根误差函数计算粒子适应度,从而选举出个体的局部最优位置和全局最优位置,然后进行迭代,若当前适应度优于个体最优值和全局最优值,则更新个体最优值和全局最优值,直到满足条件输出最优解为止。
(3)采用最小二乘法得到最小二乘解和PSO寻得的输入权重和阈值进行逐层训练,得到相关量化特征,通过DELM分类器映射到目标输出空间,得到最终的预测结果。
采用基于RF-PSO-DELM算法构建盐沼湿地碱蓬LAI反演模型。利用留出法训练模型,即取103个样方点数据作为训练集,剩下20个样方点作为测试集。为减小变量之间的差异性,反演前进行归一化处理,反演后进行相应地反归一化操作。详细反演步骤和流程如
图5 基于RF-PSO-DELM的LAI反演流程图
Fig. 5 Flow chart of LAI based on RF-PSO-DELM
目前,大部分研究没有明确提出ELM中隐藏层节点数的选取准则,依据主观性选取,不具一般性(
图6 隐含层节点数与决定系数的关系
Fig. 6 The relationship between the number of hidden layer nodes and the coefficient of determination
图7 测试集不同算法反演效果比较
Fig. 7 Comparison of inversion results of test set
从
在“Sigmoid函数”作为激活函数的模型中,对应的均方根误差在0.1左右,直接取决定系数最大时隐含层对应的节点数。因此,DELM模型选取Sigmoid作为最优的激活函数,对应的最优隐含层为20,且每一层的节点数为15。此外,利用均方根误差RMSE(Root Mean Squared Error)和模型决定系数(R2)作为碱蓬LAI反演模型精度的评价标准。
不同模态数据的影响系数即为每一个影响因素对于碱蓬LAI的影响程度(
模态数据 | 影响因素 | 影响系数 |
---|---|---|
土壤因子 | 全盐 | 0.156 |
全氮 | 0.408 | |
全磷 | 0.298 | |
全钾 | 0.073 | |
有机质 | 0.065 | |
总碳 | 0.046 | |
含水量 | 0.312 | |
红边参数 | 红边面积 | 0.197 |
红边位置 | 0.009 | |
红边振幅 | 0.438 | |
红边偏度 | 0.072 | |
红边峰度 | 0.118 | |
纹理特征 | 均值 | 0.669 |
相异性 | 0.071 | |
协同性 | 0.093 | |
二阶矩 | 0.062 | |
相关性 | 0.088 | |
对比度 | 0.126 | |
方差 | 0.153 | |
信息熵 | 0.081 | |
植被指数 | D674/D557 | 0.721 |
D695-D555 | 0.256 | |
(D605-D697)/(D605+D697) | 0.123 | |
实测理化参数 | 植被覆盖度 | 0.732 |
注: 粗体加粗部分表示影响系数大于0.1。
不同模态数据的影响因素对碱蓬LAI的影响系数如
4.2.1 LAI反演模型构建
为了检验本研究提出的基于RF-PSO-DELM算法结合多模态数据的碱蓬LAI反演模型的有效性,在样本数据相同情况下,基于SVM、BP、ELM、DELM、PSO-DELM、RF-PSO-DELM等6种算法建立黄河三角洲碱蓬LAI的反演模型。设置SVM类型为ϵ-SVR,选择RBF为核函数,惩罚系数为2.5、核函数参数为0.07且损失函数值为0.01;BP神经网络中的隐含层节点设置为15、最大训练次数为200、学习率为0.01、训练需求精度为le-5;ELM和DELM的参数设置与RF-PSO-DELM一致,选取Sigmoid作为最优的激活函数,对应的最优隐含层为20,且每一层的节点数为15;另外,PSO-DELM中关于粒子群优化算法的参数设置与RF-PSO-DELM一致,不同的是,将不同模态的所有数据全部输入到PSO-DELM模型中。
4.2.2 模型精度评价
基于6种算法构建的碱蓬叶面积指数LAI反演模型精度比较结果如
此外,在相同样本数据和输入条件下,DELM模型的性能优于ELM、BP和SVM模型。SVM模型预测性能优于ELM和BP,这主要是因为ELM和BP容易陷入局部最优导致过拟合。结合实验过程,BP神经网络相较于ELM训练时间较长,主要是由于ELM不需要迭代微调,耗时较短。因此,采用RF-PSO双优化策略可以很好地帮助DELM优选训练样本,找到最优参数,提升模型预测性能。
4.2.3 反演模型应用
将基于RF-PSO-DELM算法构建的碱蓬LAI最优反演模型应用到黄河三角洲碱蓬滩湿地,得到研究区碱蓬LAI的空间分布,结果如
图8 研究区内碱蓬LAI空间分布图
Fig. 8 The spatial distribution of the LAI of Suaeda salsa in the study area
叶面积指数较高的主要集中分布在研究区东南部,北部碱蓬稀疏区域的LAI比较低,根据实地勘察情况,研究区选定的地理位置北部区域中有潮沟经过且距离北部渤海较近,常年受潮沟中海水的高水盐侵蚀。而且结合实测的土壤样方来看,北部区域土壤样方有机质含量和全氮含量较低,含水量和盐浓度较高,这与
本研究结合区域土壤因子、植被红边参数和植被指数等光谱特征、无人机高光谱遥感影像纹理特征和地面实测的植被覆盖度等多模态数据,发展双优化策略的RF-PSO-DELM算法,构建了滨海湿地碱蓬LAI反演模型,与传统的模型构建相比,该反演模型增加了反映滨海湿地碱蓬的多模态数据,为碱蓬LAI模型反演提供了新思路和新方法。
为了验证模型的泛化性和分析不同模态数据特征对碱蓬LAI反演的影响,采用RF-PSO-DELM和传统经典且广泛使用的SVM算法应用不同模态数据构建碱蓬叶面积指数LAI的反演模型,精度比较分析结果如
模态数据 | RF-PSO-DELM | SVM | ||
---|---|---|---|---|
RMSE | R2 | RMSE | R2 | |
土壤因子 | 0.2385 | 0.6752 | 0.4383 | 0.5246 |
红边参数 | 0.2058 | 0.7282 | 0.3893 | 0.5603 |
纹理特征 | 0.1862 | 0.8057 | 0.4702 | 0.4521 |
植被指数 | 0.1694 | 0.8371 | 0.3641 | 0.7271 |
植被覆盖度 | 0.1843 | 0.8447 | 0.3571 | 0.6287 |
从
对比基于RF-PSO-DELM和SVM算法构建的碱蓬LAI反演模型精度可知:基于RF-PSO-DELM算法的LAI预测值与真实值更接近,RF-PSO-DELM算法对于非线性特征具有更好的拟合学习能力,具有更强的泛化性。植被覆盖度对碱蓬LAI的反演精度影响较大,这表明植被覆盖度可以很好地提升碱蓬LAI反演精度,这与
为了进一步说明结合多模态数据的反演模型优越性,利用RF-PSO-DELM算法结合不同模态数据的部分结果如
图9 基于RF-PSO-DELM的不同模态部分组合方式反演结果比较
Fig. 9 Comparison of inversion results of different mode combinations based on RF-PSO-DELM
(1)以滨海湿地碱蓬为研究对象,结合无人机高光谱数据,提出了一种基于RF-PSO-DELM算法,构建了滨海湿地碱蓬叶面积指数LAI的反演模型,与SVM、BP等传统模型相比,具有反演性能更高,模型更加稳定的特点。
(2)综合考虑影响滨海湿地碱蓬生长的土壤因子、纹理特征、光谱特征和植被覆盖度等多源模态数据,并利用随机森林特征优选算法提取对碱蓬叶面积指数LAI敏感的重要影响因素,有效降低了模型反演的复杂性,大大提高了模型预测的精度。
(3)利用随机森林算法进行特征选择,粒子群算法优化阈值参数,实质是采用双向特征优化策略,避免无关特征影响,减小了DELM训练过程中的运算量,提高了训练速度。
由于滨海湿地碱蓬样本数量有限,基于区域尺度的碱蓬LAI反演模型的普适性尚需进一步验证。因此,未来可针对多个区域不同生长期的碱蓬LAI分别构建反演模型,并增加湿地覆被类型,提高湿地植被LAI反演模型的普适性。在利用多模态数据构建碱蓬LAI反演模型时,可在后续研究中进一步考虑叶片叶绿素含量等实测植物参数对碱蓬LAI反演的影响。
Chen G W, Jin R J, Ye Z J, Li Q, Gu J L, Luo M, Luo Y M, Christakos G, Morris J, He J Y, Li D, Wang H W, Song L, Wang Q X and Wu J P. 2022. Spatiotemporal mapping of salt marshes in the intertidal zone of China during 1985-2019. Journal of Remote Sensing, 2022: 9793626 [DOI: 10.34133/2022/9793626] [百度学术]
Chen Z X, Ren J Q, Tang H J, Shi Y, Leng P, Liu J, Wang L M, Wu W B, Yao Y M and Hasiyuya. 2016. Progress and perspectives on agricultural remote sensing research and applications in China. Journal of Remote Sensing, 20(5): 748-767 [百度学术]
陈仲新, 任建强, 唐华俊, 史云, 冷佩, 刘佳, 王利民, 吴文斌, 姚艳敏, 哈斯图亚. 2016. 农业遥感研究应用进展与展望. 遥感学报, 20(5): 748-767 [DOI: 10.11834/jrs.20166214] [百度学术]
Cutler A, Cutler D R and Stevens J R. 2012. Random forests//Zhang C, Ma Y Q, eds. Ensemble Machine Learning. New York: Springer: 157-175 [DOI: 10.1007/978-1-4419-9326-7_5] [百度学术]
Duan B, Liu Y T, Gong Y, Peng Y, Wu X T, Zhu R S and Fang S H. 2019. Remote estimation of rice LAI based on Fourier spectrum texture from UAV image. Plant Methods, 15(1): 124 [DOI: 10.1186/s13007-019-0507-8] [百度学术]
Fang H L, Baret F, Plummer S and Schaepman-Strub G. 2019. An overview of global leaf area index (LAI): methods, products, validation, and applications. Reviews of Geophysics, 57(3): 739-799 [DOI: 10.1029/2018RG000608] [百度学术]
Gao L, Wang X F, Gu X F, Tian Q J, Jiao J N, Wang P Y and Li D. 2017a. Exploring the influence of soil types underneath the canopy in winter wheat leaf area index remote estimating. Chinese Journal of Plant Ecology, 41(12): 1273-1288 [百度学术]
高林, 王晓菲, 顾行发, 田庆久, 焦俊男, 王培燕, 李丹. 2017a. 植冠下土壤类型差异对遥感估算冬小麦叶面积指数的影响. 植物生态学报, 41(12): 1273-1288 [DOI: 10.17521/cjpe.2017.0231] [百度学术]
Gao L, Yang G J, Li C C, Feng H K, Xu B, Wang L, Dong J H and Fu K. 2017b. Application of an improved method in retrieving leaf area index combined spectral index with PLSR in hyperspectral data generated by unmanned aerial vehicle snapshot camera. Acta Agronomica Sinica, 43(4): 549-557 [百度学术]
高林, 杨贵军, 李长春, 冯海宽, 徐波, 王磊, 董锦绘, 付奎. 2017b. 基于光谱特征与PLSR结合的叶面积指数拟合方法的无人机画幅高光谱遥感应用. 作物学报, 43(4): 549-557 [DOI: 10.3724/SP.J.1006.2017.00549] [百度学术]
Guo Y K, Liu Y L, Zhang X J and Xu M. 2019. LAI inversion using radiation transfer model and random forest regression. Engineering of Surveying and Mapping, 28(6): 17-21, 29 [百度学术]
郭云开, 刘雨玲, 张晓炯, 许敏. 2019. 利用辐射传输模型和随机森林回归反演LAI. 测绘工程, 28(6): 17-21, 29 [DOI: 10.19349/j.cnki.issn1006-7949.2019.06.004] [百度学术]
Hao X L, Zhou J J, Zhang M M, Wu J, Zhang F G and Wang Y J. 2020. Study on the change of soil nutrient content of robinia pseudoacacia forest of different converted years in hilly and gully areas of Southern Shanxi Province. Forest Resources Management, 6: 105-110, 115 [百度学术]
郝小玲, 周佳佳, 张咪咪, 吴洁, 张粉果, 王永吉. 2020. 晋南丘陵沟壑区不同退耕年限刺槐林土壤养分含量变化. 林业资源管理, (6): 105-110, 115 [DOI: 10. 13466/j. cnki. lyzygl. 2020. 06. 017] [百度学术]
He W J, Han G X, Yan K, Guan B, Wang G M, Lu F, Zhou Y F, Zhang L L and Zhou L. 2021. Effects of microtopography on plant biomass and the distribution of both soil water and salinity in coastal saline-alkali land. Chinese Journal of Ecology, 40(11): 3585-3597 [百度学术]
贺文君, 韩广轩, 颜坤, 管博, 王光美, 路峰, 周英锋, 张乐乐, 周莉. 2021. 微地形对滨海盐碱地土壤水盐分布和植物生物量的影响. 生态学杂志, 40(11): 3585-3597 [DOI: 10.13292/j.1000-4890.202111.002] [百度学术]
Hernández J A, Olmos E, Corpas F J, Sevilla F and Del Río L A. 1995. Salt-induced oxidative stress in chloroplasts of pea plants. Plant Science, 105(2): 151-167 [DOI: 10.1016/0168-9452(94)04047-8] [百度学术]
Huang J F, Wang Y, Wang F M and Liu Z Y. 2006. Red edge characteristics and leaf area index estimation model using hyperspectral data for rape. Transactions of the CSAE, 22(8): 22-26 [百度学术]
黄敬峰, 王渊, 王福民, 刘占宇. 2006. 油菜红边特征及其叶面积指数的高光谱估算模型. 农业工程学报, 22(8): 22-26 [DOI: 10.3321/j.issn:1002-6819.2006.08.005] [百度学术]
Jia J, Bai J H, Wang W, Zhang G L, Wang X, Zhao Q Q and Zhang S. 2018. Changes of biogenic elements in Phragmites australis and Suaeda salsa from salt marshes in Yellow River Delta, China. Chinese Geographical Science, 28(3): 411-419 [DOI: 10.1007/s11769-018-0959-1] [百度学术]
Jiang C H, Chen Y W, Wu H H, Li W, Zhou H, Bo Y M, Shao H, Song S J, Puttonen E and Hyyppä J. 2019. Study of a high spectral resolution hyperspectral LiDAR in vegetation red edge parameters extraction. Remote Sensing, 11(17): 2007 [DOI: 10.3390/rs11172007] [百度学术]
Jing S H, Liu J Z, Chen Y J and Zhang T J. 2018. Effects of plant communities on soil phosphorus availability in coastal wetlands. Chinese Journal of Soil Science, 49(2): 392-401 [百度学术]
靖淑慧, 刘加珍, 陈永金, 张天举. 2018. 滨海湿地不同群落对土壤磷的有效性影响研究. 土壤通报, 49(2): 392-401 [DOI: 10.19336/j.cnki.trtb.2018.02.19] [百度学术]
Kanke Y, Tubaña B, Dalen M and Harrell D. 2016. Evaluation of red and red-edge reflectance-based vegetation indices for rice biomass and grain yield prediction models in paddy fields. Precision Agriculture, 17(5): 507-530 [DOI: 10.1007/s11119-016-9433-1] [百度学术]
Li C J, Sun Y Q, Fan J L, Zhai Z Z, Yang S R, Fan J W, Wang T T, Wang S J and Zhang H. 2015. Ecological significance of planting suaeda salsa in saline/alkali soils in the lop nur potash mine. Arid Zone Research, 32(6): 1160-1166 [百度学术]
李从娟, 孙永强, 范敬龙, 翟志忠, 杨司睿, 范井伟, 王婷婷, 王世杰, 张恒. 2015. 盐地碱蓬在高盐碱土环境中的生态学意义. 干旱区研究, 32(6): 1160-1166 [DOI: 10.13866/j.azr.2015.06.16] [百度学术]
Li D, Cheng T, Zhou K, Zheng H B, Yao X, Tian Y C, Zhu Y and Cao W X. 2017. WREP: a wavelet-based technique for extracting the red edge position from reflectance spectra for estimating leaf and canopy chlorophyll contents of cereal crops. ISPRS Journal of Photogrammetry and Remote Sensing, 129: 103-117 [DOI: 10.1016/j.isprsjprs.2017.04.024] [百度学术]
Li J S, Hussain T, Feng X H, Guo K, Chen H Y, Yang C and Liu X J. 2019. Comparative study on the resistance of Suaeda glauca and Suaeda salsa to drought, salt, and alkali stresses. Ecological Engineering, 140: 105593 [DOI: 10.1016/j.ecoleng.2019.105593] [百度学术]
Li Q, Zhang X Y, Ma T J, Jiao C L, Wang H and Hu W. 2021. A multi-step ahead photovoltaic power prediction model based on similar day, enhanced colliding bodies optimization, variational mode decomposition, and deep extreme learning machine. Energy, 224: 120094 [DOI: 10.1016/j.energy.2021.120094] [百度学术]
Liu F D, Mo X, Kong W J and Song Y. 2020. Soil bacterial diversity, structure, and function of Suaeda salsa in rhizosphere and non-rhizosphere soils in various habitats in the Yellow River Delta, China. Science of the Total Environment, 740: 140144 [DOI: 10.1016/j.scitotenv.2020.140144] [百度学术]
Lu J S, Chen S M, Huang W M and Hu T T. 2021. Estimation of aboveground biomass and leaf area index of summer maize using SEPLS_ELM model. Transactions of the Chinese Society of Agricultural Engineering, 37(18): 128-135 [百度学术]
陆军胜, 陈绍民, 黄文敏, 胡田田. 2021. 采用SEPLS_ELM模型估算夏玉米地上部生物量和叶面积指数. 农业工程学报, 37(18): 128-135 [DOI: 10.11975/j.issn.1002-6819.2021.18.015] [百度学术]
Luo G L, He S Q, Tan W and Qi Y J. 2022. Determination of leaf area index in Pinus massoniana plantations of different ages. Journal of Central South University of Forestry and Technology, 42(2): 55-64 [百度学术]
罗光浪, 何世强, 谭伟, 戚玉娇. 2022. 不同林龄马尾松人工林叶面积指数的测定. 中南林业科技大学学报, 42(2): 55-64 [DOI: 10.14067/j.cnki.1673-923x.2022.02.007] [百度学术]
Ma T T, Li X W, Bai J H, Ding S Y, Zhou F W and Cui B S. 2019. Four decades’ dynamics of coastal blue carbon storage driven by land use/land cover transformation under natural and anthropogenic processes in the Yellow River Delta, China. Science of the Total Environment, 655: 741-750 [DOI: 10.1016/j.scitotenv.2018.11.287] [百度学术]
Ma Y R, Lü X, Yi X, Ma L L, Qi Y Q, Hou T Y and Zhang Z. 2021. Monitoring of cotton leaf area index using machine learning. Transactions of the Chinese Society of Agricultural Engineering, 37(13): 152-162 [百度学术]
马怡茹, 吕新, 易翔, 马露露, 祁亚琴, 侯彤瑜, 张泽. 2021. 基于机器学习的棉花叶面积指数监测. 农业工程学报, 37(13): 152-162 [DOI: 10.11975/j.issn.1002-6819.2021.13.018] [百度学术]
Ma Y R, Zhang Q, Yi X, Ma L L, Zhang L F, Huang C P, Zhang Z and Lv X. 2022. Estimation of Cotton Leaf Area Index (LAI) based on spectral transformation and vegetation index. Remote Sensing, 14(1): 136 [DOI: 10.3390/rs14010136] [百度学术]
Parker G G. 2020. Tamm review: Leaf Area Index (LAI) is both a determinant and a consequence of important processes in vegetation canopies. Forest Ecology and Management, 477: 118496 [DOI: 10.1016/j.foreco.2020.118496] [百度学术]
Qi C H, Chen M, Song J and Wang B S. 2009. Increase in aquaporin activity is involved in leaf succulence of the euhalophyte Suaeda salsa, under salinity. Plant Science, 176(2): 200-205 [DOI: 10.1016/j.plantsci.2008.09.019] [百度学术]
Qu Y, Liu S H and Xie Y. 2008. Computer simulation model of fractional vegetation cover and its parameters sensitivity. Acta Agronomica Sinica, 34(11): 1964-1969 [百度学术]
瞿瑛, 刘素红, 谢云. 2008. 植被覆盖度计算机模拟模型与参数敏感性分析. 作物学报, 34(11): 1964-1969 [DOI: 10.3724/SP.J.1006.2008.01964] [百度学术]
Sha Z Y, Wang Y W, Bai Y F, Zhao Y J, Jin H, Na Y and Meng X L. 2019. Comparison of leaf area index inversion for grassland vegetation through remotely sensed spectra by unmanned aerial vehicle and field-based spectroradiometer. Journal of Plant Ecology, 12(3): 395-408 [DOI: 10.1093/jpe/rty036] [百度学术]
Su Z B, Lu Y W, Gu J T, Gao R, Ma Z and Kong Q M. 2021. Research on rice leaf area index inversion model based on improved QGA-ELM algorithm. Spectroscopy and Spectral Analysis, 41(4): 1227-1233 [百度学术]
苏中滨, 陆艺伟, 谷俊涛, 高睿, 马铮, 孔庆明. 2021. 改进的QGA-ELM算法水稻叶面积指数反演模型. 光谱学与光谱分析, 41(4): 1227-1233 [DOI: 10.3964/j.issn.1000-0593(2021)04-1227-07] [百度学术]
Sun K, Zhang J S, Zhang C X and Hu J Y. 2017. Generalized extreme learning machine autoencoder and a new deep neural network. Neurocomputing, 230: 374-381 [DOI: 10.1016/j.neucom.2016.12.027] [百度学术]
Wang L, Wang P X, Liang S L, Qi X, Li L and Xu L X. 2019. Monitoring maize growth conditions by training a BP neural network with remotely sensed vegetation temperature condition index and leaf area index. Computers and Electronics in Agriculture, 160: 82-90 [DOI: 10.1016/j.compag.2019.03.017] [百度学术]
Wang X Q, Liu S, Li Q Y and Ma K B. 2021. Classification and discrimination of surrounding rock of tunnel based on SVM of K-fold cross validation. Mining and Metallurgical Engineering, 41(6): 126-128, 133 [百度学术]
汪学清, 刘爽, 李秋燕, 马凯彬. 2021. 基于K折交叉验证的SVM隧道围岩分级判别. 矿冶工程, 41(6): 126-128, 133 [DOI: 10.3969/j.issn.0253-6099.2021.06.031] [百度学术]
Wu W B, Li J Y, Zhang Z B, Ling C J, Lin X K and Chang X L. 2018. Estimation model of LAI and nitrogen content in tea tree based on hyperspectral image. Transactions of the Chinese Society of Agricultural Engineering, 34(3): 195-201 [百度学术]
吴伟斌, 李佳雨, 张震邦, 凌彩金, 林贤柯, 常星亮. 2018. 基于高光谱图像的茶树LAI与氮含量反演. 农业工程学报, 34(3): 195-201 [DOI: 10.11975/j.issn.1002-6819.2018.03.026] [百度学术]
Xia J B, Ren J Y, Zhang S Y, Wang Y H and Fang Y. 2019. Forest and grass composite patterns improve the soil quality in the coastal saline-alkali land of the Yellow River Delta, China. Geoderma, 349: 25-35 [DOI: 10.1016/j.geoderma.2019.04.032] [百度学术]
Xie Q Y, Dash J, Huang W J, Peng D L, Qin Q M, Mortimer H, Casa R, Pignatti S, Laneve G, Pascucci S, Dong Y Y and Ye H C. 2018. Vegetation indices combining the red and red-edge spectral information for leaf area index retrieval. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(5): 1482-1493 [DOI: 10.1109/JSTARS.2018.2813281] [百度学术]
Xie Q Y, Huang W J, Cai S H, Liang D, Peng D L, Zhang Q, Huang L S, Yang G J and Zhang D Y. 2014. Comparative study on remote sensing invertion methods for estimating winter wheat leaf area index. Spectroscopy and Spectral Analysis, 34(5): 1352-1356 [百度学术]
谢巧云, 黄文江, 蔡淑红, 梁栋, 彭代亮, 张清, 黄林生, 杨贵军, 张东彦. 2014. 冬小麦叶面积指数遥感反演方法比较研究. 光谱学与光谱分析, 34(5): 1352-1356 [DOI: 10.3964/j.issn.1000-0593(2014)05-1352-05] [百度学术]
Yan G J, Hu R H, Luo J H, Weiss M, Jiang H L, Mu X H, Xie D H and Zhang W M. 2019. Review of indirect optical measurements of leaf area index: recent advances, challenges, and perspectives. Agricultural and Forest Meteorology, 265: 390-411 [DOI: 10.1016/j.agrformet.2018.11.033] [百度学术]
Yang H Y, Song J D, Zhou T, Jin G Z, Jiang F and Liu Z L. 2019. Influences of stand, soil and space factors on spatial heterogeneity of leaf area index in a spruce-fir valley forest in Xiao Hinggan Ling, China. Chinese Journal of Plant Ecology, 43(4): 342-351 [百度学术]
杨焕莹, 宋建达, 周焘, 金光泽, 姜峰, 刘志理. 2019. 林分、土壤及空间因子对谷地云冷杉林叶面积指数空间异质性的影响. 植物生态学报, 43(4): 342-351 [DOI: 10.17521/cjpe.2018.0310] [百度学术]
Yao X, Yu K Y, Yang Y J, Zeng Q, Chen Z H and Liu J. 2017. Estimation of forest leaf area index based on random forest model and remote sensing data. Transactions of the Chinese Society for Agricultural Machinery, 48(5): 159-166 [百度学术]
姚雄, 余坤勇, 杨玉洁, 曾琪, 陈樟昊, 刘健. 2017. 基于随机森林模型的林地叶面积指数遥感估算. 农业机械学报, 48(5): 159-166 [DOI: 10.6041/j.issn.1000-1298.2017.05.019] [百度学术]
Yu F H, Feng S, Zhao Y R, Wang D K, Xing S M and Xu T Y. 2020. Inversion model of chlorophyll content in japonica rice canopy based on PSO-ELM and hyper-spectral remote sensing. Journal of South China Agricultural University, 41(6): 59-66 [百度学术]
于丰华, 冯帅, 赵依然, 王定康, 邢思敏, 许童羽. 2020. 粳稻冠层叶绿素含量PSO-ELM高光谱遥感反演估算. 华南农业大学学报, 41(6): 59-66 [DOI: 10.7671/j.issn.1001-411X.202007044] [百度学术]
Yu Y H, Wang J L, Liu G J and Cheng F. 2019. Forest leaf area index inversion based on Landsat OLI Data in the Shangri-La City. Journal of the Indian Society of Remote Sensing, 47(6): 967-976 [DOI: 10.1007/s12524-019-00950-6] [百度学术]
Zhang L, Gong Z N, Wang Q W, Jin D D and Wang X. 2019. Wetland mapping of Yellow River Delta wetlands based on multi-feature optimization of Sentinel-2 images. Journal of Remote Sensing, 23(2): 313-326 [百度学术]
张磊, 宫兆宁, 王启为, 金点点, 汪星. 2019. Sentinel-2影像多特征优选的黄河三角洲湿地信息提取. 遥感学报, 23(2): 313-326 [DOI: 10.11834/jrs.20198083] [百度学术]
Zhang M, Kang G Q, Wu L F and Guan Y. 2022. A method for capacity prediction of lithium-ion batteries under small sample conditions. Energy, 238: 122094 [DOI: 10.1016/j.energy.2021.122094] [百度学术]
Zhao S S, Pu H Y and Wei H F. 2020. Effects of different treatments on seed germination and growth of Suaeda salsa. Chinese Wild Plant Resources, 39(2): 1-6 [百度学术]
赵珊珊, 蒲红宇, 魏海峰. 2020. 不同处理对碱蓬种子萌发和土壤氮磷含量的影响. 中国野生植物资源, 39(2): 1-6 [DOI: 10.3969/j.issn.1006-9690.2020.02.001] [百度学术]
Zhu T, Yu J, Xie D H and Liu L M. 2014. Particle swarm optimization in PolSAR image unspuervised classification. Journal of Geomatics Science and Technology, 31(1): 57-61 [百度学术]
朱腾, 余洁, 谢东海, 刘利敏. 2014. 粒子群优化算法在全极化SAR影像非监督分类中的应用. 测绘科学技术学报, 31(1): 57-61 [DOI: 10.3969/j.issn.1673-6338.2014.01.013] [百度学术]
相关作者
相关机构