基于多光谱荧光成像技术和SVM的葡萄霜霉病早期检测研究

葡萄霜霉病是全球危害最严重的葡萄病害,对该病进行早期检测和防治,可提高葡萄品质和产量,提出一种基于多光谱荧光成像技术（MFI）和支持向量机模型（SVM）的霜霉病早期检测方法。对人工接种霜霉病的葡萄叶片（145个）和健康对照叶片（145个）从叶背面连续6天进行多光谱荧光成像,获得试验叶片16个荧光参数（4个单独波段F440,F520,F690,F740及其相互比值）的图像。在分析不同荧光波段图像随接种天数（DPI）变化规律基础上,通过单因素方差分析和相关性分析,优选出进行霜霉病早期检测的4个波段特征F520,F690,F440/F740,F690/740,利用这4个特征构建基于SVM的霜霉病检测模型。试验发现,16个荧光参数都有早期检测霜霉病的潜力,四个单独波段中F440和F520比F690和F740对霜霉病的侵染更敏感,6DPI才显症的病斑能在F440和F520波段2DPI（接种后第二天）的荧光图像中凸显,接种叶片F440和F520波段荧光强度均随着DPI增加快速升高,在2DPI显著高于健康叶片（p<0.01）,并随着DPI增加更加显著（p<0.0001）;接种叶片F690和F740波段荧光强度均随着DPI增加逐渐减小,1DPI-3DPI与健康叶片无显著差异,从4DPI开始显著低于健康叶片（p<0.05）,并在5DPI-6DPI更加显著（p<0.01）;健康叶片荧光参数变化很小。F440极易受干扰,变异系数最大,F520最稳定。随着DPI增加,叶片被侵染程度加深,4个特征融合的SVM模型对健康和接种叶片检测准确率逐渐提高,1DPI的准确率为65.6%,3DPI检测准确率为82.2%,整个试验周期（1DPI-6DPI）的平均检测准确率达84.6%,高于单一特征中最优波段F520的阈值检测结果（1DPI的准确率为61.1%,3DPI检测准确率为78.9%,整个试验周期为80.0%）。结果表明利用MFI技术和SVM模型能实现霜霉病显症前的早期检测,为便携式葡萄霜霉病早期诊断设备的开发提供了理论依据。     

水稻叶绿素浓度对荧光光谱的影响

为了加强对水稻生长状况的监测,指导水稻的田间施肥,提高施肥利用效率,以增加农作物产量、提高粮食品质,实验室搭建了基于激光诱导荧光技术的荧光探测系统,以研究水稻叶片的叶绿素含量与荧光峰值比之间的相关性。文中测量样本为水稻分蘖期和拔节期的倒二叶,栽培地区位于中国江汉平原。文中先采用凯氏定氮法和相应公式结合测得水稻叶片的叶绿素含量(mg·g-1),再用搭建的荧光探测系统采集了水稻叶片不同叶绿素含量的荧光光谱(激发波长为355 nm)。获得了水稻叶片在不同叶绿素含量下的荧光光谱数据库,定量分析了荧光峰值比F₇₄₀/F₆₈₅(荧光谱峰740 nm、685 nm处的荧光强度比)与叶绿素含量的相关性,发现叶绿素含量的变化对荧光光谱特性影响明显。由实验数据分析可知荧光参数中的峰值比(F₇₄₀/F₆₈₅)与叶绿素含量呈现很好的线性正相关,相关系数(R2)在水稻的分蘖期和拔节期分别达到了0.901 3和0.912 5。实验分析结果表明诱导荧光光谱技术具有方便、快捷、无损等优点,在农作物生长状况的遥感定量监测等方面也具有一定发展潜力。     

氮胁迫下水稻的激光诱导荧光光谱特性

为了指导水稻的田间施肥,解决因过量施肥造成资源的大量浪费以及环境污染,特别是水体富营养化等问题,搭建了基于紫外激光诱导荧光技术的荧光光谱探测系统,以研究水稻叶片的氮水平与荧光强度的相关性。文中先用植物营养测定仪(TYS-3N)测定水稻叶片氮含量和叶绿素含量（样本为水稻分蘖期的倒二叶,采集于中国江汉平原地区）,再用搭建的测量系统采集了不同氮水平的荧光光谱。获得了不同氮水平下水稻叶片的荧光光谱数据库,分析了荧光光谱参数F₇₄₀/F₆₈₅（为峰值740,685 nm处的荧光强度比）与氮水平的相关性,发现氮含量的变化对水稻叶片荧光光谱特性影响明显。通过该实验验证了荧光参数的峰值比F₇₄₀/F₆₈₅与氮含量呈明显的线性正相关,相关系数（r）为0.871 8,均方根误差（RMSE）为0.076 82。实验表明采用激光诱导荧光光谱探测技术具有快速无损等优点,且有一定潜力用于定量测量植被营养元素的含量,为采用荧光技术对农作物施氮管理提供了支持。     

Fluorescence spectroscopy applied to orange trees

In this work, we have applied laser-induced fluorescence spectroscopy to investigate biological processes in orange trees (Citrus aurantium L.). We have chosen to investigate water stress and Citrus Canker, which is a disease caused by the Xanthomonas axonopodis pv. citri bacteria. The fluorescence spectroscopy was investigated by using as an excitation source a 442-nm 15-mW HeCd gas multimode discharge laser and a 532-nm 10-mW Nd³⁺:YAG laser. The stress manifestation was detected by the variation of fluorescence ratios of the leaves at different wavelengths. The fluorescence ratios present a significant variation, showing the possibility to observe water stress by fluorescence spectrum. The Citrus Canker’s contaminated leaves were discriminated from the healthy leaves using a more complex analysis of the fluorescence spectra. However, we were unable to discriminate it from another disease, and new fluorescence experiments are planned for the future.     

基于叶绿素荧光光谱指数的温室黄瓜病害预测

温室蔬菜病害的发生及大面积流行严重影响设施农业的生产管理, 大大降低设施农业的经济效益。为了实现温室蔬菜病害的无损准确预测, 以黄瓜霜霉病害为例, 利用激光诱导叶绿素荧光构建光谱特征指数, 建立了温室蔬菜病害的预测模型。在试验中采用对比分析的方法, 通过对作物健康叶片接种病菌孢子, 分别采集健康、接种2 d、接种6 d和出现明显病症共4组试验样本的光谱曲线, 定性分析了荧光强度随叶片样本感染病菌孢子的变化规律;利用光谱曲线不同波段峰谷值创建了叶绿素荧光光谱指数k₁=F₆₈₅/F₅₁₂和k₂=F₇₃₄/F₅₁₂, 根据数值的变化范围, 设定k₁和k₂分别为20和10时可以作为判断样本出现明显病症与未出现明显病症的特征值, 其判断的准确率分别达到96%和94%;利用构建的光谱指数与样本健康状况的分类结果, 选择光谱指数F₆₈₅/F₅₁₂,F₆₈₅-F₇₃₄,F₇₁₅/F₆₁₂可以定性判断样本健康状况, 并选择光谱指数F₆₈₅/F₅₁₂,F₇₃₄/F₅₁₂,F₆₈₅-F₇₃₄,F₇₁₅/F₆₁₂作为建立定量分析模型的输入量, 以预测集分类准确率作为评价标准, 对比判别分析、BP神经网络、支持向量机三种数据建模方法, 结果表明支持向量机作为霜霉病害预测的建模方法, 其预测能力达到91.38%。利用激光诱导叶绿素荧光构建光谱指数方法, 研究植物病害的预测问题, 具有很好的分类和鉴别效果。     

运用偏振荧光的手段研究氧化胁迫对小麦叶绿体荧光的影响

采集抽穗期小麦旗叶,采用1mmol.L-1 H2O2、干旱、黑暗处理24h诱导产生氧化损伤模型,然后运用偏振荧光的手段检测了叶绿体的荧光发射谱和荧光激发谱,结果发现,无论选择436nm激发叶绿素a(Chla)分子,或固定475nm激发叶绿素b(Chlb)分子,氧化胁迫后光系统Ⅱ反应中心P680与光系统Ⅰ反应中心P700的荧光发射峰峰面积比值A684/A720呈上升趋势;通过比较偏振荧光激发谱上E436/E475和E475/E600比值,发现随着氧化胁迫的进行,Chla对于反应中心能量传递的相对贡献大于Chlb;此外,类胡萝卜素向Chlb能量传递效率在各个偏振方向上均有所提高;通过计算偏振度及粘度,发现氧化胁迫处理促使680nm处荧光偏振度提高,内囊体膜微环境粘度增加。上述结果为研究氧化胁迫提供了一种简单、易行的方法。     

Remote Estimation of the Chlorophyll Concentration of Living Trees Using Laser-induced Fluorescence Imaging Lidar

Monthly variation in chlorophyll concentration of living ginko tree leaves 65 m away from a system was remotely estimated by a laser-induced fluorescence imaging lidar. The combination of a pulsed laser and a short-time gated CCD using an image intensifier made it possible to monitor the weak fluorescence signal from the ginko tree leaves as an image. By applying the experimental idea that a ratio of intensity of the chlorophyll fluorescence at 740 nm to that at 685 nm showed a linear correlation to the chlorophyll concentration, the fluorescence image of the ginko tree obtained by the lidar was converted to the chlorophyll concentration distribution image.     

793 nm抽运下的掺Tm3+碲酸盐玻璃的发光特性

研究了掺Tm^3 碲酸盐玻璃在793nm钛宝石激光器抽运下的发光特性，结果发现除常规的^3H4→^3H6的811nm荧光外，还存在很强的^1G4→H5的785nm二光子转换荧光以及较弱的710nm和677nm三光子上转换荧光。     

厌氧产甲烷受抑制过程的三维荧光光谱解析

应用三维荧光光谱表征了厌氧产甲烷反应受抑制前后的出水,采用平行因子分析方法分析,得到各主成分的激发发射光谱图以及相对浓度.分析结果显示,运行正常的产甲烷反应器出水以蛋白质、辅酶NADH和F420为主;产甲烷反应受到抑制后,甲烷生成速率迅速下降,同时辅酶F420荧光消失,反应器出水中以蛋白质和辅酶NADH为主.样品稀释后...     

叶绿素荧光成像技术的葡萄霜霉病早期检测

葡萄霜霉病对葡萄生产构成严重威胁,尽早防治是治理霜霉病的关键.为了对该病进行早期检测,以PCR检测获取的霜霉病相对生物量作为霜霉病侵染的依据,从暗适应—光适应—暗弛豫3个光合生理状态连续变化过程中,采集80个人工接种霜霉菌叶片和80个健康对照叶片连续6 d的叶绿素荧光图像.对比健康和接种叶片叶绿素荧光动力学曲线、参数图...     

Performance Check of Vegetation Fluorescence Imaging Lidar through In Vivo and Remote Estimation of Chlorophyll Concentration Inside Plant Leaves

Laser-induced fluorescence imaging lidar was developed for in vivo plant/vegetation monitoring. Fluorescences of poplar tree leaves growing naturally at a distance of 60 m from the lidar were successfully detected as two spectral images at wavelengths of 685 nm and 740 nm. By comparing chlorophyll concentration quantified with high performance liquid chromatography, it was confirmed that the intensity ratio of the two wavelengths was converted into the chlorophyll concentration inside the leaves. The intensity ratio of the images reflected the status of the poplar tree in the process of senescence so well that it was possible to assess the living status of the tree as a numerical value. The performance characteristics and the potential of the laser-induced fluorescence imaging lidar for monitoring the physiological activities of plants and vegetation are described.     

重金属Cu胁迫下典型农作物叶片日光诱导荧光辐射特征提取研究

典型农作物在受到重金属Cu胁迫的情况下,日光诱导荧光辐射特征会发生变化,可以间接地反映植被受重金属胁迫的强度。采用不同含量的Cu溶液或Cu污染土壤作为培养基质,选择小麦、豌豆和白菜三种典型农作物进行Cu胁迫实验,通过地物光谱仪和积分球耦合测量方式获得在不同Cu胁迫强度下的植物叶片的表观反射光谱。基于叶片荧光光谱特性,采用模型反演的方法将表观荧光光谱从表观反射光谱中提取出来,并采用三种算法(GM,AM和LM)对表观荧光光谱进行重吸收校正,获得了在不同Cu含量胁迫下典型农作物的荧光光谱,并建立了远红荧光峰(FRF)高度与叶片中Cu含量之间的正相关关系。     

色氨酸的非线性分频荧光研究

色氨酸 (Trp)在 350nm处产生一个荧光峰 ,在 70 0nm处产生一个 (分频 )荧光峰 ,此两峰荧光强度F3 50nm 和F70 0nm 均与Trp浓度 ( 0～ 1× 10 -5mol·L-1 )成线性关系 ,随着Trp浓度增大 ,350和 70 0nm半峰宽(Δλ) 3 50 ,(Δλ) 70 0 缓慢减小 ,而F70 0nm/F3 50nm 值和半峰宽比 (Δλ) 70 0 /(Δλ) 3 50 为一常数 ,此两峰具有相似的荧光特性。根据建立的分频荧光能级原理和非线性共振分频荧光原理探讨了色氨酸分频荧光峰产生的原因     

Tm:YAG晶体光谱的温度特性及荧光的温度淬灭

通过测量不同温度下Ｔｍ：ＹＡＧ晶体的吸收光谱、荧光光谱和荧光寿命发现,温度是影响Ｔｍ：ＹＡＧ晶体光谱特性的重要因素。温度升高不仅引起Ｔ：ＹＡＧ晶体吸收光谱、荧光光谱的加宽和新光谱线的产生,而且引起荧光淬灭现象的发生。     

LiF∶F2晶体中实现三维光数据存储的实验研究

基于LiF∶F2晶体在近红外飞秒脉冲激光作用下产生F2色心向F+3色心的转变,以及两种色心荧光光谱的不同,实现了荧光反射共焦读出和多光子写入的三维光数据存储的原理性实验.钛宝石再生放大器输出的脉冲宽度100 fs、中心波长800 nm、重复频率1 kHz的超短脉冲激光束,用数值孔径0.68的显微物镜聚焦到LiF：F2晶体内部,通过移动晶体实现了三维逐位式数据写入；用405 nm的连续蓝光激发存储位,通过探测F+3色心产生的540 nm荧光,实现了对信息位进行快速非破坏性的反射共焦读出.与多光子三维存储的透射共焦散射读出和相衬读出相比,格式与现存光盘技术兼容,结构简单；更重要的是,存储信息位是依靠荧光光谱的变化,折射率变化很小,可以有效增加读出层数.     

Novel Quadruple Fluorescence Properties of Two Benzoylthiourea Isomers

Two benzoylthiourea isomers, N-2-flurobenzoy-N'-4- (N,N-dimethyl)amidophenylthiourea (2FBDAPT) and N-4-fluro-benzoy-N'-4- (N,N-dimethyl)amidophenylthiourea (4FBDAPT) were determined by fourier transform infrared spectroscopy (FTIR), nuclear magnetic resonance (NMR) and X-ray diffraction. It was found that intra- and intermolecular hydrogen bonds played an important role in determining their conformations. Electronic spectra of the two compounds were investigated by UV absorption and steady-state fluorescence methods. The intermolecular hydrogen bond between the title compounds and methanol molecules caused the long wavelength absorption bands in methanol to weaken and vanish indeed. Quadruple fluorescence bands in ultraviolet and visible region were observed in the studied solvents upon the variable excitation wavelength. As same as Azumaya's suggestions for benzanilide (BA), F4 fluorescence bands with the maximum wavelength (λ(max)) between 546?nm and 622?nm were characteristic of TICT fluorescence. F3 bands of λ(max) from 434?nm to 483?nm were explained by the ESIPT model of the S1 state of the thiol tautomer to the S1 state of the keto tautomer. The new proposition was that F2 bands with λ(max) at about 365?nm were attributed to ESIPT from the S1 state of the thiol tautomer to the S0 state of the enol tautomer. And F1 fluorescence emissions with λ(max) at about 310?nm originated from the local S1 transitions of the enol tautomer. All experimental results were supported by MP2, CASSCF and CASPT2 quantum chemical calculations.     

叶绿素荧光分析技术在野生植物响应多环芳烃（菲）胁迫的应用研究

以叶绿素荧光分析为技术手段，通过研究土壤中多环芳烃（菲）胁迫对野生大豆叶绿素荧光特性以及光能分配参数的影响。结果表明：土壤菲胁迫降低了PSⅡ反应中心活性和电子传递能力，导致其光能利用能力降低。200 mg·kg-1菲胁迫促使F_v/F_m，q_P，ETR和NPQ的发生变化，降低了光合电子传递链上的电子传递能力和光合反应活性；在不同光强的调控下，土壤菲胁迫浓度的增大使得大豆幼苗叶片叶绿素荧光响应曲线Ф_PSⅡ和q_P参数呈现降低趋势，NPQ的上调启动了叶黄素循环途径耗散过剩的辐射激发能，以维持光合机构的正常生理功能。土壤菲胁迫下野生大豆叶片的F_m，F_v/F_m，F_v/Fo和PI_ABS等参数均随着菲含量的增加而降低，即土壤菲胁迫抑制了野生大豆叶片的PSⅡ光化学活性。通过对PSⅡ电子供体侧和受体侧的电子供应及传递能力的研究发现，土壤菲胁迫下野生大豆叶片OJIP曲线上0.3 ms时(即K点)的荧光强度增加，即放氧复合体(OEC)活性降低。土壤菲胁迫还导致OJIP曲线上J点和I点荧光强度的增加，说明土壤菲胁迫导致了野生大豆叶片PSⅡ受体侧电子接受能力的降低，使电子由Q_A向Q_B传递受阻。野生大豆叶片的光能吸收和分配参数也明显受到土壤菲胁迫的影响；随着土壤菲浓度的增加野生大豆叶片PSⅡ反应中心吸收光能用于Q-_A以后的电子传递能量比例和单位反应中心捕获用于电子传递的能量降低，即吸收的光能用于光化学反应的比例降低。因此，土壤菲胁迫下导致PSⅡ电子供体侧OEC的损伤和PSⅡ电子受体侧电子传递能力的降低，以及光能分配利用的改变均是导致其PSⅡ反应中心的活性降低的重要原因。研究结果表明，以叶绿素荧光分析技术可为环境中多环芳烃（菲）胁迫对植物光合作用的影响机理研究提供指导。     

发粉红色荧光的含铽络合物光谱表征

本文报道合成了可发出粉红色荧光的含铽络合物２－萘甲酸－邻菲罗啉－铽，并对其进行了荧光光谱及红外光谱研究，结果表明，由于２－萘甲酸，邻菲罗啉配体的混配作用，使铽离子的＾５Ｄ４－＾７Ｆ３跃迁产生较强的６１５ｎｍ发射峰，其发射强度大于由＾５Ｄ４－＾７Ｆ５跃迁产生的５４５ｎｍ发射峰，因而络合行发出粉红色荧光。     

Stress-Induced Alteration of Chlorophyll Fluorescence Polarization and Spectrum in Leaves of <Emphasis Type="Italic">Alocasia macrorrhiza</Emphasis> L. Schott

The value of intrinsic chlorophyll fluorescence polarization, and the intensity in emission spectrum were investigated in leaf segments of Alocasia macrorrhiza under several stress conditions including different temperatures (25–50°C), various concentrations of NaCl (0–250 mM), methyl viologen (MV, 0–25 μM), SDS (0–1.0%) and NaHSO₃ (0–80 μM). Fluorescence emission spectrum of leaves at wavelength regions of 500–800 nm was monitored by excitation at 436 nm. The value of fluorescence polarization (P value), as result of energy transfer and mutual orientation between chlorophyll molecules, was determined by excitation at 436 nm and emission at 685 nm. The results showed that elevated temperature and concentrations of salt (NaCl), photooxidant (MV), surfactant (SDS) and simulated SO₂ (NaHSO₃) treatments all induced a reduction of fluorescence polarization to various degrees. However, alteration of the fluorescence spectrum and emission intensity of F₆₈₅ and F₇₃₁ depended on the individual treatment. Increase in temperature and concentration of NaHSO₃ enhanced fluorescence intensity mainly at F₆₈₅, while an increase in MV concentration led to a decrease at both F₆₈₅ and F₇₃₁. On the contrary, NaCl and SDS did not cause remarkable change in fluorescence spectrum. Among different treatments, the negative correlation between polarization and fluorescence intensity was found with NaHSO₃ treatments only. We concluded that P value being measured with intrinsic chlorophyll fluorescence as probe in leaves is a susceptible indicator responding to changes in environmental conditions. The alteration of P value and fluorescence intensity might not always be shown a functional relation pattern. The possible reasons of differed response to various treatments were discussed.     

基于叶绿素荧光光谱分析的稻瘟病害预测模型

为了实现稻瘟病的快速、准确和无损检测,力求构建稻瘟病害预测模型。根据水稻叶片相对病害面积将稻瘟病划分为3个等级,通过激光诱导法采集不同病害等级的活体水稻叶片叶绿素荧光光谱。选取502～830 nm波段激光诱导叶绿素荧光光谱(LICF)作为研究对象,利用Savitzky-Golay平滑法(SG)和一阶导数变换(FDT)对光谱信息进行预处理,通过主成分分析(PCA)方法获取经SG-FDT预处理后光谱的特征向量,根据累积贡献率和方差选取前3个主成分进行分析。将试验样本分为建模样本和检验样本,以稻瘟病害等级为预测指标,利用建模样本的133片叶片的光谱和病害信息分别结合判别分析(DA)、多类逻辑回归分析(MLRA)和多层感知器(MLP)建立稻瘟病的预测模型,利用检验样本的89片叶片的光谱和病害信息对所建模型进行预测检验,完成对PCA-DA、PCA-MLRA和PCA-MLP的对比寻优。结果表明,PCA-DA,PCA-MLRA和PCA-MLP模型均能完成对稻瘟病害的预测,但PCA-MLP模型的平均预测准确率能够达到91.7%,相比PCA-DA和PCA-MLRA模型,在稻瘟病害3个等级上均具有较好的分类和预测能力。