基于PROSAIL辐射传输模型的毛竹林叶面积指数遥感反演

采用PROSAIL辐射传输模型建立毛竹林叶面积指数(LAI) 冠层反射率查找表,并结合Landsat TM卫星遥感数据,实现了毛竹林LAI的定量反演.结果表明： PROSAIL模型各输入参数的敏感性由高到低依次为LAI>叶绿素含量(C_ab)>叶片结构参数(N)>平均叶倾角(ALA)>等效水厚度(C_w)>干物质含量(C_m),并以LAI、C_ab两个主要敏感因子用于构建毛竹林LAI 冠层反射率查找表;基于PROSAIL模型的毛竹林LAI遥感反演结果与实测LAI具有很好的一致性,二者相关系数为0.90,均方根误差和相关的均方根误差也较小,分别为0.58和13.0%,但也存在反演LAI平均值高于实际值的问题.     

利用冠层光谱估测烟草叶面积指数和地上生物量

综合多种烟草类型、品种及肥料处理因素,分析了17种光谱参数与烟草叶面积指数（LAI）、地上鲜生物重（AFW）、地上干生物重（ADW）的关系,建立逐步回归模型对烟草LAI、AFW、ADW进行估测并结合相关分析筛选出相应的特征变量。结果表明：5个回归方程的复确定系数R^2、回归系数相伴概率均达到显著水平。包含17个光谱参量的逐步回归方程筛选出的第一自变量均为Rg／Rr,相关分析及散点图分析亦得出Rg／Rr与LAI、AFW、ADW相关系数分别为0．759、0．611、0．647,R^2为0．576、0．3727、0．4184,均达到极显著水平,证明烟草LAI、AFW、ADW的特征变量为Rg／Rr。仅采用8种植被指数建立模型,证明利用比值植被指数（RVI）估测LAI、ADW亦是可行的。经过统计检验,建立的模型估测效果均较好,估测值与实测值的相关性均达到显著水平,其中包含特征变量Rg／Rr的回归模型估测效果优于RVI构建的模型。表明采用高分辨率光谱或宽波段光谱提取光谱变量可对烟草LAI、AFW、ADW进行监测,并可根据数据条件选择有效的估测模型,为烟草遥感数据分析提供方法。     

华南地区亚热带树木叶面积指数的高光谱反演研究

为构建树种叶面积指数的估算模型,以NDVI、RVI、FREP、CIGreen、CIRed-edge、MSAVI2为高光谱特征变量,通过统计分析,确定反演树种叶面积指数的最佳光谱特征变量,构建华南农业大学校园内50种亚热带树木的叶片反射率和叶面积指数(LAI)模型。结果表明,6种高光谱特征变量与树种叶面积指数间都具有极显著相关性,其中红边位置反射率(FREP)和比值植被指数(RVI)与LAI的拟合方程的R2都大于0.8,决定系数分别为0.820和0.811。经过精度验证,FREP估算的均方根误差(RMSE)只有0.13,该回归模型为估测亚热带典型树种的叶片LAI最佳模型。从高光谱遥感的角度结合亚热带植被的群落结构特点来看,建立的红边位置光谱反射率与叶面积指数的回归模型普遍具有较高的拟合度,所以利用高光谱特征变量反演亚热带树木叶片的叶面积指数等植被参数的应用前景较好。     

利用数字植物冠层图象分析仪测定南亚热带森林叶面积指数的初步报道

介绍了当今较先进的数字植物冠层图象分析仪的理论基础、实际应用的特点 ,测定了南亚热带丘陵地区 8种森林类型 (包括人工林和自然次生林 )的叶面积指数、太阳直射透过系数等群落冠层特征 ,同时测定一由荒坡草地自然发育而成的灌木草坡群落的叶面积指数。根据测定数据比较分析了不同人工林群落 1 1a发展这些特征的变化 ,并与次生自然林的测定数据进行比较 ,同时还对 1 1a与 1 4a树龄马占相思林林冠变化进行比较分析。     

林地叶面积指数遥感估算方法适用分析

叶面积指数是与森林冠层能量和CO2交换密切相关的一个重要植被结构参数,为了探讨估算林地叶面积指数LAI的遥感适用方法和提高精度的途径,利用TRAC仪器测定北京城区森林样地的LAI,从Landsat TM遥感图像计算NDVI、SR、RSR、SAVI植被指数,分别建立估算LAI的单植被指数统计模型、多植被指数组合的改进BP神经网络,获取最有效描述LAI与植被指数非线性关系的方法并应用到TM图像估算北京城区LAI。结果表明,单植被指数非线性统计模型估算LAI的精度高于线性统计模型;多植被指数组合神经网络中,以NDVI、RSR、SAVI组合估算LAI的精度最高,估算值与观测值线性回归方程的R2最高,为0.827,而RMSE最低,为0.189,神经网络解决了多植被指数组合统计模型非线性回归方程的系数较多、较难确定的问题,可较为有效的应用于遥感图像林地LAI的估算。     

基于叶面积指数反演的区域冬小麦单产遥感估测

利用定量遥感技术反演的叶面积指数（LAI）在中国北方黄淮海地区典型县市进行冬小麦单产预测研究.为提高数据质量和减少估产误差,利用Savitzky-Golay滤波技术降低云对NDVI数据的影响及数据缺失;通过冬小麦实测LAI进行时序内插,模拟得到实测点每日冬小麦LAI,继而获得实测点主要生育时期平均LAI;在此基础上,建立了冬小麦主要生育时期平均LAI与作物单产关系模型,改变目前利用生育时期内某一时间点LAI代替整个生育时期LAI的方法;在模型择优基础上,得到最佳遥感估产关键期--开花期LAI与单产统计模型;最后,利用MODIS-NDVI经验模型反演得到的开花期平均LAI进行2008年冬小麦单产预测.结果表明：与地面实测的冬小麦单产相比,研究区估产平均相对误差为1.21%,RMSE达到257.33 kg·hm^-2,可以满足大范围估产的要求.利用上述方法可以在研究区冬小麦收获前20～30 d进行准确的单产估计.     

多源多角度遥感数据反演森林叶面积指数方法

利用北京1号和Landsat多源数据组合成4个角度多波段数据集,在考虑森林三维垂直分布特点的基础上,结合INFORM几何光学与辐射传输混合模型,通过聚类＋神经元网络方式,建立相应的多源多角度LAI反演模型。最后利用实地LAI测量数据和MODISLAI产品,对不同角度组合和噪声水平下的LAI反演结果进行验证。结果表明：在保证数据质量的条件下,通过增加角度可以提高森林的LAI反演精度,最终R2=0.713,RMSE=0.957,比单个角度的反演精度平均提高约20%。     

多源多角度遥感数据反演森林叶面积指数方法

利用北京1号和Landsat多源数据组合成4个角度多波段数据集, 在考虑森林三维垂直分布特点的基础上, 结合INFORM几何光学与辐射传输混合模型, 通过聚类+神经元网络方式, 建立相应的多源多角度LAI反演模型。最后利用实地LAI测量数据和MODIS LAI产品, 对不同角度组合和噪声水平下的LAI反演结果进行验证。结果表明: 在保证数据质量的条件下, 通过增加角度可以提高森林的LAI反演精度, 最终R²=0.713, RMSE=0.957, 比单个角度的反演精度平均提高约20%。     

利用数字图像估测棉花叶面积指数

叶面积指数是指示棉花长势、产量形成和高产群体调控等信息的重要结构特征参数。本研究的目的在于利用基于冠层图像光照叶片和光照土壤分量的图像透光率估测棉花叶面积指数。通过3年不同种植密度、品种、施氮量和灌水量的田间试验,在棉花不同的生育期用数码相机、LAI-2000冠层仪和线性光量子传感器采集数据并进行破坏性取样测定,分析图像透光率的有效性和建立LAI估测模型,进而对图像方法、LAI-2000和破坏性取样方法进行比较和分析。结果表明：(1)在太阳高度角最大且变化最小的正午时段,数码相机测量的图像透光率与线性光量子传感器测量的冠层透光率较一致且相对稳定。(2)图像透光率能反映除吐絮期以外各时期的冠层透光状况,但是当LAI大于5时图像透光率出现饱和。(3)综合分析2009和2010年数据,建立了图像透光率估测LAI的模型(R2=0.8438, SE=0.5605);利用2007年独立试验资料检验估测模型的性能,模型检验的拟合度较高(R2=0.8767)且预测误差较小(RMSE=0.4305),当LAI＞5时模型的预测能力降低。(4)数字图像、LAI-2000和破坏性取样三种方法测量的LAI值之间均呈现显著的线性相关(R2＞0.85),但是图像透光率的饱和性致使当LAI＞5时明显低估叶面积指数。     

湿地松林叶面积指数测算

对湿地松（Pinus elliotii）当年生和多年生两针一束、三针一束叶片的长度、宽度、厚度和重量分别进行量测,据此探讨不同类型叶片的叶形和比叶面积差异,并结合样地调查数据对中科院千烟洲试验站湿地松人工林的叶面积指数进行计算。结果表明：湿地松三针一束叶片合拢后横切面基本呈圆形,当年生叶和多年生叶的平均直径分别为1．688mm和1.706mm;两针一束叶片合拢后从统计学上讲横切面不是圆形,而是椭圆形,叶片厚度方向直径大于宽度方向（当年生叶厚度和宽度方向直径分别为1．580ram和1．422mm,多年生叶分别为1．568mm和1.410mm）,但如果把厚度和宽度方向直径的平均值近似成圆柱体直径计算时误差在3％以内;如果只用厚度或宽度方向直径代表平均直径计算结果会有2％。10％的误差;当年生叶和多年生叶、两针一束叶和三针一束叶之间比叶面积差别很大,计算的三种比叶面积（投影比叶面积、圆柱面比叶面积和比表面积）中,当年生叶的比叶面积明显大于多年生叶,三针一束叶片的投影比叶面积和比表面积都大于两针一束叶片,但圆柱面比叶面积恰好相反。湿地松林的叶面积指数若按投影叶面积算为3．61,按圆柱面的外表面算为5．12,按总表面积的一半算为4．52,比利用冠层分析仪测量的结果略大。     

1988-2013年基于BP神经网络的植被叶面积指数遥感定量反演

叶面积指数（Leaf Area Index,LAI）高度综合了植被水平覆盖状况和垂直结构,以及枯枝落叶层厚薄和地下生物量多少,是植被影响土壤侵蚀的主要方面。区域尺度的时间序列叶面积指数揭示了区域土壤侵蚀的演化过程。因此,及时准确地掌握区域尺度上长时间序列的植被LAI,对研究土壤侵蚀动态变化与植被的关系至关重要。选择南京市1988-2013年10期遥感影像,基于反向传播（Back Propagation,BP）神经网络构建LAI反演模型,进行了长时间序列的叶面积指数反演。结合2009和2010年LAI实测值,验证与探讨了该模型的评价精度与适应性。结果表明：（1）该模型拟合度较高,2009和2010年平均相对误差、均方根误差、相关系数分别是0.2395和0.2174,0.2962和0.2581,0.7713和0.6844,各项精度评价指标均较好;（2）统计分析去除耕地后全市LAI变化,低植被覆盖（LAI<2）面积不断增加,高植被覆盖区（LAI>3）面积先减少后增加,耕地面积不断减少,符合南京市的发展变化规律;（3）主城区LAI年际变化与其他学者得到的南京市植被盖度变化趋势一致,反演结果的时序性较高。本文提出的基于反向传播神经网络模型反演长时间序列LAI是可行的,为区域尺度土壤侵蚀定量遥感监测提供新途径。     

基于PROSAIL辐射传输模型的毛竹林叶面积旨数遥感反演

采用PROSAIL辐射传输模型建立毛竹林叶面积指数(LAI)-冠层反射率查找表,并结合Landsat TM卫星遥感数据,实现了毛竹林LAI的定量反演.结果表明:PROSAIL模型各输入参数的敏感性由高到低依次为LAI＞叶绿素含量(Cab)＞叶片结构参数(N)＞平均叶倾角(ALA)＞等效水厚度(Cw)＞干物质含量(Cm),并以LAI、Cab两个主要敏感因子用于构建毛竹林LAI-冠层反射率查找表;基于PROSAIL模型的毛竹林LAI遥感反演结果与实测LAI具有很好的一致性,二者相关系数为0.90,均方根误差和相关的均方根误差也较小,分别为0.58和13.0％,但也存在反演LAI平均值高于实际值的问题.     

Recognition method of soybean leaf diseases using residual neural network based on transfer learning

Leaf disease is an important factor restricting the high quality and high yield of the soybean plant. Insufficient control of soybean diseases will destroy the local ecological environment and break the stability of the food chain. To overcome the low accuracy in recognizing soybean leaf diseases using traditional deep learning models and complexity in chemical analysis operations, in this study, a recognition model of soybean leaf diseases was proposed based on an improved deep learning model. First, four types of soybean diseases (Septoria Glycines Hemmi, Soybean Brown Leaf Spot, Soybean Frogeye Leaf Spot, and Soybean Phyllosticta Leaf Spot) were taken as research objects. Second, image preprocessing and data expansion of original images were carried out using image registration, image segmentation, region calibration and data enhancement. The data set containing 53, 250 samples was randomly divided into the training set, verification set, and test set according to the ratio of 7:2:1. Third, the convolution layer weight of the pre-training model based on the ImageNet open data set was transferred to the convolution layer of the ResNet18 model to reconstruct the global average pooling layer and the fully connected layer for constructing recognition model of TRNet18 model. Finally, the recognition accuracy of the four leaf diseases reached 99.53%, the Macro-F1 was 99.54%, and the average recognition time was 0.047184 s. Compared with AlexNet, ResNet18, ResNet50, and TRNet50 models, the recognition accuracy and Macro-F1 of the TRNet18 model were improved by 6.03% and 5.99% respectively, and the model recognition time was saved by 16.67%, The results showed that the proposed TRNet18 model had higher classification accuracy and stronger robustness, which can not only provide a reference for accurate recognition of other crop diseases, but also be transplanted to the mobile terminal for recognition of crop leaf diseases.     

Automated classification of urinary stones based on microcomputed tomography images using convolutional neural network

PurposeThe classification of urinary stones is important prior to treatment because the treatments depend on three types of urinary stones, i.e., calcium, uric acid, and mixture stones. We have developed an automatic approach for the classification of urinary stones into the three types based on microcomputed tomography (micro-CT) images using a convolutional neural network (CNN).Materials and methodsThirty urinary stones from different patients were scanned in vitro using micro-CT (pixel size: 14.96 μm; slice thickness: 15 μm); a total of 2,430 images (micro-CT slices) were produced. The slices (227 × 227 pixels) were classified into the three categories based on their energy dispersive X-ray (EDX) spectra obtained via scanning electron microscopy (SEM). The images of urinary stones from each category were divided into three parts; 66%, 17%, and 17% of the dataset were assigned to the training, validation, and test datasets, respectively. The CNN model with 15 layers was assessed based on validation accuracy for the optimization of hyperparameters such as batch size, learning rate, and number of epochs with different optimizers. Then, the model with the optimized hyperparameters was evaluated for the test dataset to obtain classification accuracy and error.ResultsThe validation accuracy of the developed approach with CNN with optimized hyperparameters was 0.9852. The trained CNN model achieved a test accuracy of 0.9959 with a classification error of 1.2%.ConclusionsThe proposed automated CNN-based approach could successfully classify urinary stones into three types, namely calcium, uric acid, and mixture stones, using micro-CT images.     

基于叶面积指数估算植被总初级生产力

长时间序列的陆地碳通量数据在全球生态环境变化研究中具有重要意义。采用MODIS GPP(Gross Primary Productivity)算法,基于GIMMS LAI3g,MODIS15和Improved-MODIS15三种叶面积指数(LAI),估算了全球2000至2010年的植被总初级生产力(GPP)。该估算的GPP数值经过全球20个通量站点的验证,并结合MODIS17分析了它们在时空变化上的异同。结果表明:(1)4种GPP精度如下:GPP_(MOD17)GPP_(impro_MOD15)GPP_(LAI3g)GPP_(MOD15)。(2)4种GPP整体上具有一致的季节波动,冬季和夏季整体好于春季和秋季。GPP_(LAI3g)的4个季节精度较相近,而GPP_(MOD17)除了春秋季外其它季节都较好。(3)GPP_(LAI3g)在中等GPP值分布区的估值相对较高,其全球总GPP大体为(117±1.5)Pg C/a,GPP_(MOD17)和GPP_(impro_MOD15)相近且都低于该值。(4)GPP_(LAI3g)和GPP_(impro_MOD15)在大约63.29%的陆面上呈显著(P0.05)的正相关关系,它们和GPP_(MOD17)在LAI不确定性小的地区呈显著的正相关关系。GPP_(LAI3g)和GPP_(MOD15)正相关分布面积占比为40.61%。     

The computation of forest leaf area index on slope using fish-eye sensors

Leaf area index (LAI) measured with fish-eye sensors is inevitably affected by slope. In order to quantify this effect, LAIs measured in contrasted forests were corrected for slope. Digitised hemispherical photographs were analysed to compute LAI according to azimuth and as a function of slope. The topographic mask (part of hemisphere obstructed by slope) and the path length of a ray of light at a given zenith angle (longer upslope and shorter downslope) were taken into account. Results indicate a strong slope effect, both in temperate conifer forest plantation and in tropical rain forest. On regular slope, this effect is easily corrected, although the topographic mask affects data. On irregular topography, corrected and uncorrected LAIs strongly differ, suggesting a need for reference measurements; corrected figures appear nevertheless realistic and normality is reached only for corrected values. LAIs should systematically be corrected for slope above 15-20 degrees, more so if topography is irregular or the forest cover heterogeneous.     

基于数码相片Gamma校正的水稻叶面积指数估算

随着数码相机的日益普及,利用数码相机进行作物叶面积指数(LAI)测量不断得以应用。由于数码相机成像时会对入射光辐射强度进行Gamma编码变换,输出的相片DN(Digital Number)值与入射光辐射强度呈非线性关系,会造成在确定相片中植被叶片与背景的分割阈值时出现误差,并最终导致LAI估算存在较大不确定性。以水稻为研究对象,获取不同生长期水稻冠层相片并结合同步LAI 2000测量的LAI数据,基于相片Gamma校正原理,对水稻不同生长期冠层相片进行Gamma校正,在此基础上利用冠层孔隙率方法,估算不同生长期水稻LAI。结果表明,经过Gamma校正相片估算的水稻LAI总体精度有显著提高,相片估算的IMAGE LAI与LAI-2000测量值比较的决定系数达到0.71(P0.05)。在整个观测期内,两种方法观测的LAI值在时间变化趋势上表现一致,但在不同生长期内存在差别,在水稻分蘖期和拔节期相片估算的IMAGE LAI要高于LAI-2000测量值,孕穗期到抽穗期期间IMAGE LAI低于LAI-2000测量值,乳熟期到成熟期IMAGE LAI又高于LAI-2000的观测结果。     

SCAPEVIEWER: preliminary results of a landscape perception classification system based on neural network technology

In this paper, the implementation of a pilot computerized system for the classification of landscape images (SCAPEVIEWER) is presented. A total of 108 landscape photographs have been organized, according to the mean estimation of scenic beauty from seven experts, into three classes: indistinctive (C1), typical or common (C2), and distinctive (C3). For each of the landscape photographs, 10 indices are estimated. These indices are then fed to a classifier based on neural network (NN) technology. In order to examine whether NNs are suitable for this specific application, two different approaches have been tested and compared against a linear discrimination method (LDM) classifier. The first approach is a feed forward NN (Classic-NN), while the second approach (Hybrid-NN) is based on the Classic-NN modified by using genetic algorithms (GAs). The correct classification performances achieved by the Classic-NN and the Hybrid-NN were 87% and 84%, respectively, while the classification performance of the LDM classifier was only 68%. Although the Classic-NN achieved slightly better results than the Hybrid-NN, the latter is preferred due to its ability of index selection and automatical adjustment of internal NN parameters. The pilot system has shown the feasibility for classifying landscape photographs according to scenic beauty by means of a computerized system combining the knowledge of an expert with a NN classifier.