A New Discrimination Method of Maize Seed Varieties Based on Near-Infrared Spectroscopy

A new discrimination method for the maize seed varieties based on the near-infrared spectroscopy was proposed. The reflectance spectra of maize seeds were obtained by a FT-NIR spectrometer (12 000-4 000 cm(-1)). The original spectra data were preprocessed by first derivative method. Then the principal component analysis (PCA) was used to compress the spectra data. The principal components with the cumulate reliabilities more than 80% were used to build the discrimination models. The model was established by Psi-3 neuron based on biomimetic pattern recognition (BPR). Especially, the parameter of the covering index was proposed to assist to discriminating the variety of a seed sample. The authors tested the discrimination capability of the model through four groups of experiments. There were 10, 18, 26 and 34 varieties training the discrimination models in these experiments, respectively. Additionally, another seven maize varieties and nine wheat varieties were used to test the capability of the models to reject the varieties not participating in training the models. Each group of the experiment was repeated three times by selecting different training samples at random. The correct classification rates of the models in the four-group experiments were above 91. 8%. The correct rejection rates for the varieties not participating in training the models all attained above 95%. Furthermore, the performance of the discrimination models did not change obviously when using the different training samples. The results showed that this discrimination method can not only effectively recognize the maize seed varieties, but also reject the varieties not participating in training the model. It may be practical in the discrimination of maize seed varieties.

土丘陵沟壑区植被恢复过程中的土壤抗蚀与细沟侵蚀演变

对黄土丘陵沟壑区安塞纸坊沟和县南沟、延安燕沟3个流域不同恢复年限的植物群落的土壤抗蚀性和侵蚀程度进行了研究。对12个土壤抗蚀性指标进行主成分分析表明,土壤抗蚀性(主成分综合指数)强弱为灌木群落阶段>多年生草本和蒿类群落阶段>一二年生草本群落阶段,与一二年生草本群落阶段相比,灌木群落阶段与多年生草本和蒿类群落阶段的土壤抗蚀性分别增加了362.29%～673.33%和574.71%～930.00%;野外调查结果分析表明,随着植被的恢复演替,土壤侵蚀量呈现明显的下降趋势,灌木群落阶段的土壤侵蚀量仅为演替初期的1.42%～5.59%;通过回归分析,土壤侵蚀量和水稳性团聚类因子,以及有机质含量之间分别存在极显著与显著相关关系,鉴于土壤分析的易获性,可选择>0.5mm水稳性团聚体与有机质含量作为反映土壤侵蚀程度的指标。

环境因素影响草地柠条生长的主成分及典型相关分析

Principal Component and Canonical Correlation Analysis of the Environmental Factors Influencing the Growth of Caragana korshinskii Kom. in Grassland

黄土高原子午岭地区植被恢复过程中的土壤质量评价

选择黄土高原子午岭地区植被自然恢复1~140 a不同年限的阳坡坡地为研究对象,通过相关性分析筛选土壤表层(0~20 cm)16项表征土壤物理、化学、生物学性质的指标,运用主成分分析计算土壤质量综合指数,评价子午岭地区植被恢复过程对土壤质量的影响。结果表明:土壤总孔隙度、平均重量直径(MWD)、有机质质量分数、速效磷质量分数、蔗糖酶活性、碱性磷酸酶活性、真菌数量、微生物总量构成土壤质量评价指标体系;土壤质量综合指数随植被恢复年限的增加而增加;植被自然恢复1~140 a间,土壤质量综合指数变化范围为0.155 9~0.850 9,较裸露休闲地增加4.2~27.5倍;根据土壤质量综合指数变化规律,可将140 a植被恢复过程中的土壤质量演变过程分为3个阶段,即植被恢复初期(1~20 a)的土壤质量综合指数呈快速增长,植被恢复中期(20~40 a)的土壤质量综合指数呈波动性增长,植被恢复后期(40~140 a)的土壤质量综合指数呈稳定增长。植被演替过程中不同植被生活型土壤质量综合指数表现为乔木林地>灌木林地>草地。

陕北风沙滩地区紫花苜蓿地下滴灌带埋设深度初步研究

研究了半干旱沙区不同滴灌带埋设深度下紫花苜蓿的生长特性。通过试验研究分析了滴灌带埋设深度对紫花苜蓿植株高度、茎粗、分枝数、根系生长、根系密度和产量等生长特性的影响。采用主成分分析法对不同滴灌带埋设深度的紫花苜蓿等生长特性进行了综合评价。结果表明,滴灌带不同埋设深度对苜蓿各个生育期生长特性指标影响不同。在苗期,埋设深度为10 cm的处理,有利于苜蓿生长。从分枝期起,埋设深度为30 cm的处理优于其它处理;在整个生育期内,不同埋设深度对苜蓿生长特性影响的综合评判结果为:埋深30 cm>埋深20 cm>埋深10 cm>埋深40 cm。

PCA在语音检测中的应用研究

提出主元分析PCA(Principal Component Analysis)用于语音检测的方法研究．用主元分析法在多维空间中建立坐标轴，将待处理信号投影到该坐标轴中，通过分析投影结果判断是否为语音信号．通过将语音和非语音分别建立子空间，来区分语音和非语音信号．该方法不同于常规的语音时域、频域处理方法，而是在多维空间中对信号进行分析·实验结果表明，该方法准确率高、简单、容易实现，而且能区分多种非语音信号．

岷江干旱河谷典型阳坡土壤质量动态与评估

横断山区干旱河谷地区的土壤是制约植被恢复的一个关键要素，但干旱条件下土壤质量与动态演变状况并不清楚，本研究以岷江干旱河谷核心地段的土壤为对象，研究了阳坡褐土土壤的物理、化学以及生物学特征在海拔梯度（1800m、2000m、2200m）格局及其时间动态（2005-2007）变化，应用主成分分析综合评价了土壤质量及其时空差异性，明确了土壤质量的变化趋势。主要研究结论如下： 1) 海拔梯度上土壤物理性状的变化，2005 年和2007 年土壤物理性状综合质量随着海拔的升高均得到了优化，即海拔2200 m＞海拔2000 m＞海拔1800 m。 2) 海拔梯度上土壤化学性质的变化，土壤化学综合性质2005 年随海拔升高而趋于变得更优，而2007 年海拔2000 m 最优，海拔1800 m 则最差。 3) 海拔梯度上土壤生物学性质的变化，2005 年土壤生物学性质随海拔升高表现出趋于更好，2007 年海拔2000 m 最优而海拔1800 m 地段最差。 4) 从土壤物理、化学、生物学三方面出发，应用主成分分析，分别分析得出2005 年和2007 年不同海拔高度的土壤质量综合得分。根据综合得分得出土壤质量综合评价的排列顺序为：2005 海拔2200 m＞2007 年海拔2000 m＞2005 年海拔2000 m＞2007 年海拔1800 m＞2007 海拔2200 m＞2005 年海拔1800 m。2005年土壤综合质量随海拔升高而趋好，2007 年则以海拔2000 m 最优，海拔1800 m和2200 m 差异不大。 5）排除人为干扰后，干旱河谷土壤物理性状在海拔1800 m 略有恢复，海拔2000 m 变化不明显，而海拔2200 m 仍有退化趋势；土壤化学性质在海拔1800 m和2000 m 地段得到恢复，而海拔2200 m 处仍有退化；土壤生物学性质在海拔2000 m 地段呈恢复趋势，而1800 m 和2200 m 仍处于退化状态。综合质量分析表明，与2005 年相比，2007 年海拔1800 m 和2000 m 地段土壤质量趋于变优而海拔2200 m 地段仍有退化迹象。 Soil is a key factor that affect the restoration of vegetation in the Hengduan Mountains dry valley area. But the dynamics and quality of soil is not knowed in dry area. In this study, soil physiochemical and biological characteristics ranging from 1800~2200m above sea level from a typical south-facing slope at the Minjiang River dry valley area had been studied, and characteristics of changes in soil quality along altitudinal gradients and time scales were also discussed. The principal component analysis was used to assess the soil quality. The main results were as follows: 1) Changes in soil physical properties along altitudes. Soil physical properties obtained the optimization along with the elevation in 2005 and 2007. 2) Changes in soil chemical properties. It was summarized that soil chemical properties increased with elevation in 2005, but the soil of 2000 m was the best in 2007. 3) Changes in soil biological characteristics. Soil biological properties increased with elevation in 2005, but the soil of 2000 m was the best and 1800 m was the worst in 2007. 4) Change tendency of soil quality. With the soil physics, chemistry and biology characteristics, we analysised the change tendency of soil quality in altitudes. The result indicated that soil quality increased with altituded in 2005, and soil quality of 2000 m was the best in 2007. 5) In brief, the soil quality is by physics, chemistry as well as the biology synthesizes the influence the final outcome. And the soil quality's change was manifested by soil physics, chemistry and the biology characteristics. All the results indicated soil quality still degenerated at 2200 m in in the dry valley of Minjiang River. And soil quality of 1800 m and 2000 m resumed slight.

Artificial neural network classification based on high-performance liquid chromatography of urinary and serum nucleosides for the clinical diagnosis of cancer

Nucleosides in human urine and serum have frequently been studied as a possible biomedical marker for cancer, acquired immune deficiency syndrome (AIDS) and the whole-body turnover of RNAs. Fifteen normal and modified nucleosides were determined in 69 urine and 42 serum samples using high-performance liquid chromatography (HPLC). Artificial neural networks have been used as a powerful pattern recognition tool to distinguish cancer patients from healthy persons. The recognition rate for the training set reached 100%. In the validating set, 95.8 and 92.9% of people were correctly classified into cancer patients and healthy persons when urine and serum were used as the sample for measuring the nucleosides. The results show that the artificial neural network technique is better than principal component analysis for the classification of healthy persons and cancer patients based on nucleoside data. (C) 2002 Elsevier Science B.V. All rights reserved.

A QUANTITATIVE ASSESSMENT OF SOURCE CONTRIBUTIONS TO INHALABLE PARTICULATE MATTER POLLUTION IN METROPOLITAN BOSTON

In this paper, source apportionment techniques are employed to identify and quantify the major particle pollution source classes affecting a monitoring site in metropolitan Boston, MA. A Principal Component Analysis (PCA) of paniculate elemental data allows the estimation of mass contributions for five fine mass panicle source classes (soil, motor vehicle, coal related, oil and salt aerosols), and six coarse panicle source classes (soil, motor vehicle, refuse incineration, residual oil, salt and sulfate aerosols). Also derived are the elemental characteristics of those source aerosols and their contributions to the total recorded elemental concentrations (i.e. an elemental mass balance). These are estimated by applying a new approach to apportioning mass among various PCA source components: the calculation of Absolute Principal Component Scores, and the subsequent regression of daily mass and elemental concentrations on these scores.

Spatial variability in water quality and surface sediment diatom assemblages in a complex lake basin,lake of the woods,ontario,canada

Lake of the Woods (LOW) is an international waterbody spanning the Canadian provinces of Ontario and Manitoba, and the U.S. state of Minnesota. In recent years, there has been a perception that water quality has deteriorated in northern regions of the lake, with all increase in the frequency and intensity of toxin-producing cyanobacterial blooms. However, given the lack of long-term data these trends are difficult to verify. As a first step, we examine spatial and seasonal patterns in water quality in this highly complex lake on the Canadian Shield. Further, we examine surface sediment diatom assemblages across multiple sites to determine if they track within-take differences in environmental conditions. Our results show that there are significant spatial patterns in water quality in LOW. Principal Component Analysis divides the lake into three geographic zones based primarily on algal nutrients (i.e., total phosphorus, TP), with the highest concentrations at sites proximal to Rainy River. This variation is closely tracked by sedimentary diatom assemblages, with [TP] explaining 43% of the variation in diatom assemblages across sites. The close correlation between water quality and the surface sediment diatom record indicate that paleoecological models could be used to provide data on the relative importance of natural and anthropogenic sources of nutrients to the lake.

小麦抗旱生态分类中适合性聚类方法的研究

探索了适合于小麦品种抗旱生态分类的聚类方法 .选用 2 1个农艺性状和 15个冬小麦品种 (系 ) ,在聚类分析的各环节上 ,通过采用不同的策略 ,大规模进行了各种分类结果的比较 .结果表明 ,在与专家经验分类接近程度上 ,数据转换方法中 ,原始数据法依次大于普通相关阵基础上的方差极大正交旋转法、Promax斜交旋转法、主成份法 ;相似性度量上 ,欧氏距离大于马氏距离 ;聚类方式上 ,对应分析法和模糊聚类法大于最短距离法、最长距离法、类平均法 ;所有可组合的方法中 ,以对应分析法和直接用原始数据的模糊聚类法的分类结果最接近专家经验分类 .结合各方法理论上优缺点的分析与检验 ,认为这两种方法也是较理想的方法 .

Fast Haar Transform Based Feature Extraction for Face Representation and Recognition

Subspace learning is the process of finding a proper feature subspace and then projecting high-dimensional data onto the learned low-dimensional subspace. The projection operation requires many floating-point multiplications and additions, which makes the projection process computationally expensive. To tackle this problem, this paper proposes two simple-but-effective fast subspace learning and image projection methods, fast Haar transform (FHT) based principal component analysis and FHT based spectral regression discriminant analysis. The advantages of these two methods result from employing both the FHT for subspace learning and the integral vector for feature extraction. Experimental results on three face databases demonstrated their effectiveness and efficiency.

L1-Norm-Based 2DPCA

In this paper, we first present a simple but effective L1-norm-based two-dimensional principal component analysis (2DPCA). Traditional L2-norm-based least squares criterion is sensitive to outliers, while the newly proposed L1-norm 2DPCA is robust. Experimental results demonstrate its advantages.

Supervised Gaussian Process Latent Variable Model for Dimensionality Reduction

The Gaussian process latent variable model (GP-LVM) has been identified to be an effective probabilistic approach for dimensionality reduction because it can obtain a low-dimensional manifold of a data set in an unsupervised fashion. Consequently, the GP-LVM is insufficient for supervised learning tasks (e. g., classification and regression) because it ignores the class label information for dimensionality reduction. In this paper, a supervised GP-LVM is developed for supervised learning tasks, and the maximum a posteriori algorithm is introduced to estimate positions of all samples in the latent variable space. We present experimental evidences suggesting that the supervised GP-LVM is able to use the class label information effectively, and thus, it outperforms the GP-LVM and the discriminative extension of the GP-LVM consistently. The comparison with some supervised classification methods, such as Gaussian process classification and support vector machines, is also given to illustrate the advantage of the proposed method.