Rapid Determination of Oxygen Saturation and Vascularity for Cancer Detection

A rapid heuristic ratiometric analysis for estimating tissue hemoglobin concentration and oxygen saturation from measured tissue diffuse reflectance spectra is presented. The analysis was validated in tissue-mimicking phantoms and applied to clinical measurements in head and neck, cervical and breast tissues. The analysis works in two steps. First, a linear equation that translates the ratio of the diffuse reflectance at 584 nm and 545 nm to estimate the tissue hemoglobin concentration using a Monte Carlo-based lookup table was developed. This equation is independent of tissue scattering and oxygen saturation. Second, the oxygen saturation was estimated using non-linear logistic equations that translate the ratio of the diffuse reflectance spectra at 539 nm to 545 nm into the tissue oxygen saturation. Correlations coefficients of 0.89 (0.86), 0.77 (0.71) and 0.69 (0.43) were obtained for the tissue hemoglobin concentration (oxygen saturation) values extracted using the full spectral Monte Carlo and the ratiometric analysis, for clinical measurements in head and neck, breast and cervical tissues, respectively. The ratiometric analysis was more than 4000 times faster than the inverse Monte Carlo analysis for estimating tissue hemoglobin concentration and oxygen saturation in simulated phantom experiments. In addition, the discriminatory power of the two analyses was similar. These results show the potential of such empirical tools to rapidly estimate tissue hemoglobin in real-time spectral imaging applications.     

Assessment of embryonic growth in chicken eggs by means of visible transmission spectroscopy

During this work, it was investigated whether spectral measurements can be used to monitor embryonic growth. An experiment was conducted in which both the transmission spectra and embryonic weight were determined on 240 eggs (Cobb, 37 weeks) between Day 5 and Day 10 of incubation. The spectral data were linked to embryonic weight by means of a partial least squares analysis. Different preprocessing procedures were compared during this work, that is, smoothing, multiplicative scatter correction (MSC), and first‐ and second‐order derivative. Compared to the remainder of the preprocessing procedures, MSC leads to a considerable improvement of the prediction capability of the embryonic weight. The ratio of performance to deviation obtained for the MSC spectra equaled 4.5 indicating that a very accurate prediction of embryonic weight is feasible based on the VIS/NIR transmission measurements. Important regions for the prediction are situated around 685–740 nm. It is suggested that the spectral changes in these spectral regions result from the displacement of carotenoids from the yolk into the blood circuitry. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2010     

On the validity of the spectrophotometric determination of oxygen saturation of hemoglobin. The wavelength dependence of observed oxygen equilibrium parameter values

Spectral changes of oxyhemoglobin induced by such anions as 2,3-diphosphoglycerate, inositol hexaphosphate, and Cl- may affect the validity of the spectrophotometric determination of oxygen saturation of hemoglobin. Therefore, the anion-induced difference spectra were extensively measured under a variety of conditions and accurate oxygen equilibrium curves were determined under representative conditions with detection at different wavelengths selected from peaks, troughs, and zero difference points of the difference spectra in the visible and Soret regions. Oxygen equilibrium parameters including the four Adair constants (i.e., equilibrium constants for four steps of oxygenation) estimated from the equilibrium curves did not show any dependence on wavelength within the limits of experimental error. These results indicate that anion-induced spectral changes do not invalidate the spectrophotometric determination of oxygen saturation and confirm the validity of the previous conclusions drawn in our series of studies on the effects of anions, pH and temperature on oxygen equilibrium parameters.     

不同水氮条件下水稻冠层反射光谱与植株含水率的定量关系

研究了不同土壤水氮条件下水稻 (Oryzasativa) 冠层光谱反射特征和植株水分状况的量化关系。结果表明, 水稻冠层近红外光谱反射率随土壤含水量的降低而降低, 短波红外光谱反射率随土壤含水量的降低而升高。相同土壤水分条件下, 高氮水稻的冠层含水率高于低氮水稻的冠层含水率 ;同一水分条件下, 高氮处理的可见光区和短波红外波段光谱反射率低于低氮处理, 近红外波段光谱反射率高于低氮处理。发现拔节后比值植被指数 (R810 /R460 ) 与水稻叶片含水率和植株含水率呈极显著的线性相关, 模型的检验误差 (RootmeansquareError, RMSE) 分别为 0.93和 1.5 0。表明比值植被指数R810 /R460 可以较好地监测不同生育期水稻叶片和植株含水率。     

Sensitivity of ex situ and in situ spectral surface plasmon resonance sensors in the analysis of protein arrays

We have investigated the sensitivity of ex situ (analysis under air condition) and in situ (analysis under liquid condition) spectral SPR sensors, which were self-constructed with fiber optic spectrometers. The sensitivity of SPR sensors was analyzed in the wavelength range of 550-780 nm by the interactions of streptavidin and biotinylated IgG, and the sensitivity was dependent on the wavelength of measurements. The sensitivity of an ex situ SPR sensor operated at the long wavelength range from 712 nm was approximately 2.6 times higher than that at the short wavelength range from 571 nm. In addition, the sensitivity of an ex situ spectral SPR sensor was about twice as high as that of an in situ spectral SPR sensor for the same resonance wavelength range. This was interpreted in that the difference in sensitivity between two SPR sensors was significantly caused by the evanescent field intensity at the metal/dielectric interface. Thus, it was suggested that ex situ spectral SPR sensors operated at the long wavelength range are sensitive biosensors for the high-throughput analysis of protein interactions on protein arrays.     

Use of dual wavelength spectrophotometry and continuous enzymatic depletion of oxygen for determination of the oxygen binding constants of hemoglobin

A small stopped-flow cuvette was built into a computer-controlled Cary 210 spectrophotometer. The enzymatic depletion of oxygen in solutions of hemoglobin and myoglobin was initiated by flowing the hemeproteins with the enzyme against a solution of the hemeproteins containing the appropriate substrate. The deoxygenation was homogeneous throughout the solution. Oxygen activity was calculated at each instant of time from the fractional saturation of Mb, determined from observations at the Hb/HbO2 isosbestic wavelength. Fractional saturation of Hb was determined from absorbances at the Mb/MbO2 isosbestic wavelength. The spectrophotometer cycled between these two wavelengths during the deoxygenation. The deoxygenation of HbO2 was largely complete in 20-25 min, whereas the deoxygenation of MbO2 was allowed to proceed for about 1 h. This procedure eliminates equilibration of Hb solutions with a gas phase and replaces oxygen electrode readings with spectrophotometric sensing by Mb, providing essentially instantaneous determinations of oxygen activity and hence 250-500 or more independent data points per run. The Mb and Hb data vectors require several manipulations to correct for small relative displacements in time and for small non-isosbestic effects. Detailed consideration of the enzyme kinetics allowed oxygen activities to be determined in regions where Mb is a poor sensor. Studies of HbO2 deoxygenation as a function of wavelength show that the determination of the four Adair constants requires in addition the determination of three spectroscopic parameters. Values of the apparent Adair constants, determined without these spectroscopic parameters, depend strongly on the monitoring wavelength.     

基于挺水植物高光谱信息的再生水总氮含量估测——以北京市门城湖湿地公园为例

高光谱信息是探测植物体内氮素含量状况的重要手段,而植物体中的氮素与水体含氮量息息相关.本研究区为以再生水为主要补给水水源的北京门城湖湿地公园,通过获取区内典型的再生水氮净化挺水植物芦苇和香蒲叶片的高光谱数据,并在室内测定对应样点的水体总氮含量指标, 探讨基于典型湿地挺水植物高光谱数据对水体总氮进行遥感探测的可行性.采用4种高光谱参数(光谱指数、归一化差值指数、“三边”参数及吸收特征参数)分别建立一元线性模型、逐步多元回归模型和偏最小二乘模型,根据决定系数(R²)和均方根误差(RMSE)进行模型精度检验.结果表明: 逐步多元回归和偏最小二乘模型的预测精度高于一元线性模型. 3种模型对芦苇的拟合效果均优于香蒲.偏最小二乘模型对芦苇的拟合效果最优(R²=0.854,RMSE=0.647).500～700 nm是反映水氮含量的最佳波段范围,绿峰与红谷反射率的比值与水体总氮含量具有较强的相关性,尤其是吸收特征参数能够较好地预测水体总氮含量.     

Spectral discrimination of coral reef benthic communities

Effective identification and mapping of coral reef benthic communities using high-spatial and -spectral resolution digital imaging spectrometry requires that the different communities are distinguishable by their spectral reflectance characteristics. In Kaneohe Bay, Oahu, Hawaii, USA, we collected in situ a total of 247 spectral reflectances of three coral species (Montipora capitata, Porites compressa, Porites lobata), five algal species (Dictyosphaeria cavernosa, Gracilaria salicornia, Halimeda sp., Porolithon sp., Sargassum echinocarpum) and three sand benthic communities (fine-grained carbonate sand, sand mixed with coral rubble, coral rubble). Major reflectance features were identified by peaks in fourth derivative reflectance spectra of coral (at 573, 604, 652, 675 nm), algae (at 556, 601, 649 nm) and sand (at 416, 448, 585, 652, 696 nm). Stepwise wavelength selection and linear discriminant function analysis revealed that spectral separation of the communities is possible with as few as four non-contiguous wavebands. These linear discriminant functions were applied to an airborne hyperspectral image of a patch reef in Kaneohe Bay. The results demonstrate the ability of spectral reflectance characteristics, determined in situ, to discriminate the three basic benthic community types: coral, algae and sand. Accepted: 12 January 2000     

Predicting leaf gravimetric water content from foliar reflectance across a range of plant species using continuous wavelet analysis

Leaf water content is an important variable for understanding plant physiological properties. This study evaluates a spectral analysis approach, continuous wavelet analysis (CWA), for the spectroscopic estimation of leaf gravimetric water content (GWC, %) and determines robust spectral indicators of GWC across a wide range of plant species from different ecosystems. CWA is both applied to the Leaf Optical Properties Experiment (LOPEX) data set and a synthetic data set consisting of leaf reflectance spectra simulated using the leaf optical properties spectra (PROSPECT) model. The results for the two data sets, including wavelet feature selection and GWC prediction derived using those features, are compared to the results obtained from a previous study for leaf samples collected in the Republic of Panamá (PANAMA), to assess the predictive capabilities and robustness of CWA across species. Furthermore, predictive models of GWC using wavelet features derived from PROSPECT simulations are examined to assess their applicability to measured data. The two measured data sets (LOPEX and PANAMA) reveal five common wavelet feature regions that correlate well with leaf GWC. All three data sets display common wavelet features in three wavelength regions that span 1732-1736 nm at scale 4, 1874-1878 nm at scale 6, and 1338-1341 nm at scale 7 and produce accurate estimates of leaf GWC. This confirms the applicability of the wavelet-based methodology for estimating leaf GWC for leaves representative of various ecosystems. The PROSPECT-derived predictive models perform well on the LOPEX data set but are less successful on the PANAMA data set. The selection of high-scale and low-scale features emphasizes significant changes in both overall amplitude over broad spectral regions and local spectral shape over narrower regions in response to changes in leaf GWC. The wavelet-based spectral analysis tool adds a new dimension to the modeling of plant physiological properties with spectroscopy data.     

On-line cell mass monitoring of Saccharomyces cerevisiae cultivations by multi-wavelength fluorescence

The catalyst in bioprocesses, i.e. the cell mass, is one of the most challenging and important variables to monitor in bioprocesses. In the present study, cell mass in cultivations with Saccharomyces cerevisiae was monitored on-line with a non-invasive in situ placed sensor measuring multi-wavelength culture fluorescence. The excitation wavelength ranged from 270 to 550 nm with 20 nm steps and the emission wavelength range was from 310 to 590 nm also with 20 nm steps. The obtained spectra were analysed chemometrically with the multi-way technique, parallel factor analysis (PARAFAC), resulting in a decomposition of the multivariate fluorescent landscape, whereby underlying spectra of the individual intrinsic fluorophors present in the cell mass were estimated. Furthermore, gravimetrically determined cell mass concentration was used together with the fluorescence spectra for calibration and validation of multivariate partial least squares (PLS) regression models. Both two- and three-way models were calculated, the models behaved similarly giving root mean square error of prediction (RMSEPs) of 0.20 and 0.19 g l(-1), respectively, when test set validation was used. Based on this work, it is evident that on-line monitoring of culture fluorescence can be used for estimation of the cell mass concentration during cultivations.     

稻麦叶片氮积累量与冠层反射光谱的定量关系

作物氮素积累动态是评价作物群体长势及估测产量和品质的重要指标,对于作物氮素的实时监测和精确管理具有重要意义。该文以5个小麦(Triticum aestivum)品种和3个水稻(Oryza sativa)品种在不同施氮水平下的3年田间试验为基础,综合研究了稻麦叶片氮积累量与冠层反射光谱的定量关系。结果表明,不同试验中拔节后叶片氮积累量均随施氮水平呈上升趋势;稻麦冠层光谱反射率在不同施氮水平下存在明显差异,可见光区(460~710 nm)反射率一般随施氮水平的增加逐渐降低,近红外波段(760~1 220 nm)反射率却随施氮水平的增加逐渐升高;就单波段而言,810和870 nm处的冠层光谱反射率均与稻麦叶片氮积累量具有相对较高的相关性;在光谱参数中,比值植被指数(Ratio vegetation index, RVI)(870,660)和RVI(810,660)均与稻麦叶片氮积累量具有高度的相关性,且相关系数明显高于单波段反射率,尤其是水稻作物;对于小麦和水稻,均可以利用统一的波段和光谱指数来监测其叶片氮积累量,并可以采用统一的回归方程来描述其叶片氮积累量随单波段反射率和反射光谱参数的变化模式,但若采用单独的回归系数则可以提高稻麦叶片氮积累量估测的准确性。     

基于高光谱的苹果花磷素含量监测模型

在室内条件下,利用ASD FieldSpec 3地物光谱仪,测定了盛花期苹果花的高光谱反射率,在分析苹果花原始光谱反射率及其一阶导数特征的基础上,分析了高光谱反射率与磷素含量间的相关关系,确定其敏感波段,构建了特征光谱参数,并建立了苹果花磷素含量的监测模型.结果表明：苹果花磷素含量与350～370 nm、670～1385 nm、1620～1760 nm波段的原始光谱反射率以及波段500～520 nm光谱反射率的一阶导数呈极显著正相关,与波段670～730 nm一阶导数呈极显著负相关;光谱参数DVI(936,676)、DVI(977,676)、NDVI(936,676)、NDVI(977,676)与苹果花磷素含量的相关性较好,其相关系数均达到0.77以上,由此建立了以4种光谱参数为自变量的磷素含量监测模型,其中,以NDVI(936,676)为自变量构建的监测模型具有最大的决定系数（R²=0.9385）、最小的均方根误差（RMSE=0.6883）和最小的相对误差（RE=7.6%）,预测精度达到92.4%,为最佳监测模型.     

植被生化组分光谱模型抗土壤背景能力分析

应用LOPEX'93(Leaf Optical Properties Experiment)数据,分析了统计回归模型在进行植被叶绿素和水分反演中抗土壤背景影响的能力,模型参数分别使用了:反射率及其变化形式、光谱位置变量、植被指数。在LOPEX'93数据库的植被波谱中分别加入10%-90%的实测土壤光谱信息,得到植被与土壤的混合光谱,并分析混合光谱对植被生化组分的响应。结果表明:应用反射率及其变化形式进行植被叶绿素反演时,以730nm和400nm组合的反射率和反射率倒数的对数为参数的模型具有最高的抗土壤背景能力,在土壤背景所占比例从低到高的变化过程中,以二者反射率组合为参数的模型与叶绿素的相关系数,始终保持在0.645附近,以二者反射率倒数的对数为参数的模型与植被叶绿素的相关系数保持在0.650附近;应用反射率及其变化形式进行植被含水量反演时,以1100,1170,1000,1040,1080nm组合的反射率为参数的模型以及以1170,960,1210,1090,1080,950,1220,1210nm反射率倒数的对数组合为参数的模型具有较高的稳定性,在土壤组分变化的过程中,以上模型与植被含水量的相关系数均稳定的高于0.99;对于光谱位置变量的分析中,以红边-绿峰-红谷组合的模型与植被叶绿素含量具有较高、而且稳定的相关系数,在土壤背景所占比例变化的情况下,相关系数稳定在0.53附近;在应用植被指数进行叶绿素的反演过程中,植被指数与叶绿素的相关系数在土壤背景所占比例变化的情况下变化较大,抗土壤背景的能力均较差;在应用植被指数进行植被水分含量的反演时,以水分指数Ratio975和Ratio1200相关系数最高,且在不同比例土壤背景变化下稳定,相关系数分别分布在0.980附近和0.960附近。该结果可用于指导不同植被覆盖条件下植被冠层参数的反演,提高反演的稳定性和准确性。     

Colorimetric device for measurement of transvascular fluid flux in blood-perfused organs

The aim of this study was to develop a device capable of measuring transvascular fluid flux in blood-perfused organs. For any given blood flow through the organ (QT), transvascular flux (QF) can be considered as the fraction of QT exchange. Presumably, QF would change the background concentration of an impermeable tracer residing in the perfusate. Thus QF could be calculated from the relative changes in tracer concentration for any given QT. We have used Blue Dextran (1 g/l of blood) as the reference tracer. Because the minimum molecular weight of Blue Dextran is 2 X 10(6), we anticipated it to behave as an impermeable tracer in most organs. QF was simulated with continuous infusions of plasma, normal saline solution, and a 50% mixture of both. Changes in Blue Dextran concentration were continuously followed colorimetrically by changes in transmission of specific light at a wavelength of 632 nm. Because 632-nm light is affected by hematocrit and O2 saturation changes, two additional wavelengths were used: 815-nm, which is not affected by saturation or Blue Dextran concentration changes, was used to account for changes in hematocrit, and 887-nm specific light, which is not affected by Blue Dextran, served to correct for saturation changes. Red cells could not be used as the reference tracer because of the possibility of hematocrit changes independent of fluid flux (Fahraeus effect). The device so constructed proved capable of measuring rates of fluid infusion in the order of 0.1% of QT with a variability of 10% around the mean.     

Estimation of leaf nitrogen content from spectral characteristics of rice canopy

Ground-based remotely sensed reflectance spectra of hyperspectral resolution were monitored during the growing period of rice under various nitrogen application rates. It was found that reflectance spectrum of rice canopy changed in both wavelength and reflectance as the plants developed. Fifteen characteristic wavebands were identified from the apparent peaks and valleys of spectral reflectance curves, in accordance with the results of the first-order differentiation, measured over the growing season of rice. The bandwidths and center wavelengths of these characteristic wavebands were different among nitrogen treatments. The simplified features by connecting these 15 characteristic wavelengths may be considered as spectral signatures of rice canopy, but spectral signatures varied with developmental age and nitrogen application rates. Among these characteristic wavebands, the changes of the wavelength in band 11 showed a positive linear relationship with application rates of nitrogen fertilizer, while it was a negative linear relationship in band 5. Mean reflectance of wavelengths in bands 1, 2, 3, 5, 11, and 15 was significantly correlated with application rates. Reflectance of these six wavelengths changed nonlinearly after transplanting and could be used in combination to distinguish rice plants subjected to different nitrogen application rates. From the correlation analyses, there are a variety of correlation coefficients for spectral reflectance to leaf nitrogen content in the range of 350-2400 nm. Reflectance of most wavelengths exhibited an inverse correlation with leaf nitrogen content, with the largest negative value (r = -0.581) located at about 1376 nm. Changes in reflectance at 1376 nm to leaf nitrogen content during the growing period were closely related and were best fitted to a nonlinear function. This relationship may be used to estimate and to monitor nitrogen content of rice leaves during rice growth. Reflectance of red light minimum and near-infrared peak and leaf nitrogen content were correlated nonlinearly.     

不同生长状态珊瑚光谱特征

珊瑚礁生态系统迅速退化是目前重要的生态环境问题之一,应用遥感技术监测大范围珊瑚礁的结构组成和变迁有很大的潜力。珊瑚光谱响应特征受珊瑚生态习性影响,在光学上相似而容易造成混淆误判。采集了西沙群岛大量石珊瑚样品的光谱,对其光谱特征进行分析及成因探讨。通过导数光谱、主成分分析研究了不同生长状态珊瑚的光谱差异,并建立珊瑚生长状态高光谱遥感判别准则。结果表明,珊瑚的光谱特性及其变化均较为复杂,受珊瑚种类和生长环境影响,光谱形状主要由共生藻色素吸收决定的。结合520—530 nm、564—574 nm和600—605 nm的导数光谱可以区分健康珊瑚、白化珊瑚和藻类覆盖的死珊瑚。总体判定准确度优于80%,误判的主要来源是种内珊瑚反射率差异。研究表明珊瑚礁环境高光谱遥感可以定量评估珊瑚状态的变化。     

Spectral response of the seagrass Zostera noltii with different sediment backgrounds

The efficiency of vegetation indices (VIs) to estimate the above-ground biomass of the seagrass species Zostera noltii Hornem. from remote sensing was tested experimentally on different substrata, since terrestrial vegetation studies have shown that VIs can be adversely influenced by the spectral properties of soils and background surfaces. Leaves placed on medium sand, fine sand and autoclaved fine sand were incrementally removed, and the spectral reflectance was measured in the 400–900 nm wavelength range. Several VIs were evaluated: ratios using visible and near infrared wavelengths, narrow-band indices, indices based on derivative analysis and continuum removal. Background spectral reflectance was clearly visible in the leaf reflectance spectra, showing marked brightness and spectral contrast variations for the same amount of vegetation. Paradoxically, indices used to minimize soil effects, such as the Soil-Adjusted Vegetation Index (SAVI) and the Modified second Soil-Adjusted Vegetation Index (MSAVI₂) showed a high sensitivity to background effects. Similar results were found for the widely used Normalized Difference Vegetation Index (NDVI) and for Pigment Specific Simple Ratios (PSSRs). In fact, background effects were most reduced for VIs integrating a blue band correction, namely the modified specific ratio (mSR₍₇₀₅₎), the modified Normalized Difference (mND₍₇₀₅₎), and two modified NDVIs proposed in this study. However, these indices showed a faster saturation for high seagrass biomass. The background effects were also substantially reduced using Modified Gaussian Model indices at 620 and 675 nm. The blue band corrected VIs should now be tested for air-borne or satellite remote sensing applications, but some require sensors with a hyperspectral resolution. Nevertheless, this type of index can be applied to analyse broad band multispectral satellite images with a blue band.     

Discrimination of coral reflectance spectra in the Red Sea

Benthic populations can potentially be mapped from remotely acquired spectral imagery, provided that they have distinctive reflectance signatures. We examined the spectral reflectance characteristics of 14 genera of Red Sea coral using a submersible spectroradiometer. Coral spectra varied quantitatively and qualitatively over the depth interval 5-20 m. Tissue pigment content had a larger effect on reflectance than colony morphology. Ten coral genera could be discriminated with a statistical probability of 52% on the basis of their absolute reflectance. Six groups of two to three coral genera could be discriminated with a probability of 60% on the basis of their rates of change in reflectance at specific wavelengths. All coral genera could be discriminated with a minimum probability of 59% on the basis of their proportionate reflectance at one wavelength as compared with another. Ratio analysis holds significant potential for interpretation of remotely sensed imagery because measurement of reflectance at six wavelengths generates 15 different parameters that can be used to identify and discriminate the component classes. The method distinguishes spectra on the basis of their shape rather than their amplitude, which helps to factor out light-induced variation in pigment concentration, exposing taxon-specific differences in pigment composition and morphology.     

滨海盐土土壤水分的高光谱参数及估测模型

基于滨海盐土5个试验点的土壤含水量和室内土壤表面高光谱反射率,综合分析了350～2500 nm波段范围内土壤含水量与土壤光谱之间的关系,并基于比值光谱指数(RSI)、归一化光谱指数(NDSI)和差值光谱指数(DI)确定了光谱参数,进而构建土壤含水量估测定量模型.结果表明： 滨海盐土原始光谱反射率与土壤含水量呈显著负相关关系,且最大负相关出现在1930 nm(r=0.86)附近.对RSI、NDSI和DI的直线回归方程、幂函数回归方程进行对比,以RSI(R₁₄₀₇,R₁₄₅₉)为自变量构建的土壤含水量指数函数线性回归方程决定系数最大(0.780),标准误较小(0.016),拟合方程为y=0.00001e^9.72053x.估测模型能够更好地监测滨海盐土土壤水分状况.基于RSI(R₁₄₀₇,R₁₄₅₉)构建的模型可实现对江苏省滨海盐土土壤水分的精确监测.     

黄土丘陵沟壑区土壤养分含量的高光谱预测模型

基于高光谱数据快速准确估算土壤养分含量,可为土壤养分监测及土壤理化参数反演提供优化方法.本研究在陕北黄土丘陵沟壑区选取典型样地,分析土壤养分含量与光谱反射率的定量关系,采用连续投影算法提取其光谱特征波长,利用偏最小二乘法、多元线性回归法、支持向量机法分别对土壤有机质、全氮、全磷、全钾含量进行预测并对比分析,构建该区域土壤养分含量的最优高光谱预测模型.结果表明: 黄土丘陵沟壑区土壤养分含量与光谱反射率在可见光区(400～760 nm)和近红外区(760～1100 nm)相关性较高,相关系数最大值均位于这两个光谱区间.4种土壤养分含量的SPA-SVM模型的普适性好且反演精度高,建模过程简单高效,适用于小数据量试验.本研究结果可为采用机器学习算法构建黄土丘陵沟壑区土壤养分含量高光谱预测模型提供参考.