Principal Component Loadings Research Articles

Assessment of riparian vegetation characteristics in Mediterranean headwaters regulated by check dams using multivariate statistical techniques

In mountainous torrents of the Mediterranean environment the riparian vegetation is strongly influenced by the presence of engineering control works, since these structures bring heavy modifications in channel geometry, hydraulic regime and bed sediment size. Previous investigations have shown high linear correlations between physical (section shape, profile slope, specific discharge, surface and subsurface size of the channel bed) and vegetation (development, structure and biodiversity) indicators in headwater channels with check dams of Calabrian (Southern Italy) torrents. Based on these findings, this study applies multivariate statistical techniques (Principal Component Analysis and Partial Least Square Regression) to identify in the same study headwaters new synthetic explanatory variables, representative of the different transects (upstream, downstream or intermediate, compared to the check dam location) and develop predictive models of riparian vegetation characteristics.The Principal Component Analysis has provided a simple parameter (the first Principal Component, explaining about 60% of the total variance), which is able to discriminate the physical and vegetal characteristics of the different transects close to check dams, thus reducing the large number of factors influencing the fluvial processes. Moreover, cover, height and transversal variability of riparian vegetation have a very high influence (loadings over 0.73) on this component, while its biodiversity is correlated to the second Principal Component (loadings over 0.63). The Partial Least Square Regression has shown that it is possible to estimate with fair accuracy (minimum r2 of 0.70) the development, structure as well as transversal variability of the riparian vegetation, starting from the physical features of the channel. These models may be important in the planning steps of new check dams, since their effects on the development and growth of vegetation upstream and downstream can be forecasted before their installation, at least for the quantification of the order of magnitude of the check dam impacts on torrent ecology.

The Same Story or a Unique Novel? Within-Participant Principal-Component Analysis of Measures of Training Load in Professional Rugby Union Skills Training.

To identify which combination metrics of external and internal training load (TL) capture similar or unique information for individual professional players during skills training in rugby union using principal-component (PC) analysis. TL data were collected from 21 male professional rugby union players across a competitive season. This included PlayerLoad™, total distance, and individualized high-speed distance (>61% maximal velocity; all external TL) obtained from a microtechnology device (OptimEye X4; Catapult Innovations, Melbourne, Australia) that was worn by each player and the session rating of perceived exertion (RPE) (internal TL). PC analysis was conducted on each individual to extract the underlying combinations of the 4 TL measures that best describe the total information (variance) provided by the measures. TL measures with PC loadings (PCL) above 0.7 were deemed to possess well-defined relationships with the extracted PC. The findings show that from the 4 TL measures, the majority of an individual's TL information (first PC: 55-70%) during skills training can be explained by session RPE (PCL: 0.72-0.95), total distance (PCL: 0.86-0.98), or PlayerLoad (PCL: 0.71-0.98). High-speed distance was the only variable to relate to the second PC (PCL: 0.72-1.00), which captured additional TL information (+19-28%). Findings suggest that practitioners could quantify the TL of rugby union skills training with one of PlayerLoad, total distance, or session RPE plus high-speed distance while limiting omitted information of the TL imposed during professional rugby union skills training.

Adaptive predictive principal components for modeling multivariate air pollution.

Air pollution monitoring locations are typically spatially misaligned with locations of participants in a cohort study, so to analyze pollution-health associations, exposures must be predicted at subject locations. For a pollution measure like PM2.5 (fine particulate matter) comprised of multiple chemical components, the predictive principal component analysis (PCA) algorithm derives a low-dimensional representation of component profiles for use in health analyses. Geographic covariates and spatial splines help determine the principal component loadings of the pollution data to give improved prediction accuracy of the principal component scores. While predictive PCA can accommodate pollution data of arbitrary dimension, it is currently limited to a small number of pre-selected geographic covariates. We propose an adaptive predictive PCA algorithm, which automatically identifies a combination of covariates that is most informative in choosing the principal component directions in the pollutant space. We show that adaptive predictive PCA improves the accuracy of multi-pollutant exposure predictions at subject locations.

Biochemical changes in injured sciatic nerve of rats after low-level laser therapy (660nm and 808nm) evaluated by Raman spectroscopy.

The aim of this study was to identify biochemical changes in sciatic nerve (SN) after crush injury and low-level laser therapy (LLLT) with 660nm and 808nm by Raman spectroscopy (RS) analysis. A number of 32 Wistar rats were used, divided into four groups (control 1, control 2, LASER 660nm, and LASER 808nm). All animals underwent surgical procedure of the SN and groups control 2, LASER 660nm, and LASER 808nm were submitted to SN crush damage (axonotmesis). The LLLT in the groups LASER 660nm and LASER 808nm was applied daily for 21 consecutive days (100mW, 30s, 133J/cm2 fluence). The hind paw was removed and the SN was dissected and positioned on an aluminum support to collect dispersive Raman spectra (830nm excitation, 30s accumulation). To estimate the biochemical changes in the SN associated with LLLT, the principal component analysis (PCA) was applied. The Raman spectra of the sciatic nerve fragments showed peaks of the major biochemical components of the nerve, especially sphingolipids, phospholipids, glycoproteins, and collagen. The spectral features identified in some of the principal component loading vectors are referred to the biochemical elements present on the SN and were increased in the groups treated with LLLT, mainly lipids (sphingo and phospholipids) and proteins (collagen)-constituents of the myelin sheath. The RS was effective in identifying the biochemical differences in the SN after the crush injury, and LASER 660nm was more efficient than the LASER 808nm in cell proliferation and repair of the injured SN.

Principal Component Approximation and Interpretation in Health Survey and Biobank Data

Background Increasing numbers of variables in surveys and administrative databases are created. Principal component analysis (PCA) is important to summarize data or reduce dimensionality. However, one disadvantage of using PCA is the interpretability of the principal components (PCs), especially in a high-dimensional database. By analyzing the variance distribution according to PCA loadings and approximating PCs with input variables, we aim to demonstrate the importance of variables based on the proportions of total variances contributed or explained by input variables. Methods There were five data sets of various sizes used to understand the performance of PC approximation: Hitters, SF-12v2 subset of the 2004 to 2011Medical Expenditure Panel Survey (MEPS), and the full set of 1996 to 2011 MESP data, along with two data sets derived from the Canadian Health Measures Survey (CHMS): a spirometry subset with the measures from the first trial of spirometry and a full data set that contained non-redundant variables. The variables in data sets were first centered and scaled before PCA. PCs approximation was studied with two approaches: PCA loadings and PC approximation through forward regression. First, the PC loadings were squared to estimate the variance contribution by variables to PCs. The other method was to use forward-stepwise regression to approximate PCs with all input variables. Results The first few PCs had large variances in each data set. Approximating PCs using stepwise regression could efficiently identify the input variables that explain large portions of PC variances than approximating according to PCA loadings in the data sets. It required fewer numbers of variables to explain more than 80% of the PC variances through stepwise regression. Conclusion Approximating and interpreting PCs with stepwise regression is highly feasible. PC approximation is useful to 1) interpret PCs with input variables, 2) understand the major sources of variances in data sets, 3) select unique sources of information and 4) search and rank input variables according to the proportions of PC variance explained. This can be an approach to systematically understand databases and search for variables that are important to databases.

Alternative butanol/gasoline and butanol/diesel fuel blends: An analysis of the interdependence between physical-chemical properties by a multivariate principal component analysis model

The use of butanol as an oxygenated component in blends with fossil fuels has recently been recognized by the industry as a promising and green alternative for automotive use, being subject of several recent studies. In this work, the interdependence between important physical-chemical properties of butanol/gasoline and butanol/diesel fuel blends was investigated using a multivariate principal component analysis model. The model dataset was based on laboratorial results of density, kinematic viscosity, distillation, vapor pressure, octane rating, anti-knock index, flash point and cetane number in a total of 48 blends, the variables of which were transformed to principal component analysis matrix representations, pre-processed and then analyzed. A good coherence was observed between the experimental results in laboratory and those derived from the principal component analysis models, evidencing important physical-chemical changes in blends’ properties due to the butanol addition. Principal component analysis scores and loadings plots could provide an intuitive and comprehensive data visualization. Butanol/gasoline fuel blends showed an overall increase in density, octane rating and higher distillation temperatures from the initial boiling point to T60 (temperature of the 60% distilled volume) and reduction of the distillation temperatures from T70 to the final boiling point. An absolute reduction in values of all properties was observed for butanol/diesel fuel blends, especially for initial distillation temperatures from initial boiling point to T35, T98, final boiling point and flash point, whereas the reductions for density, kinematic viscosity and cetane number were less intense. Total variances of up to 92.50% and 94.14% were explained by the proposed principal component analysis model, depending on the blends matrix and butanol isomer composition.

Edge-group sparse PCA for network-guided high dimensional data analysis.

Principal component analysis (PCA) has been widely used to deal with high-dimensional gene expression data. In this study, we proposed an Edge-group Sparse PCA (ESPCA) model by incorporating the group structure from a prior gene network into the PCA framework for dimension reduction and feature interpretation. ESPCA enforces sparsity of principal component (PC) loadings through considering the connectivity of gene variables in the prior network. We developed an alternating iterative algorithm to solve ESPCA. The key of this algorithm is to solve a new k-edge sparse projection problem and a greedy strategy has been adapted to address it. Here we adopted ESPCA for analyzing multiple gene expression matrices simultaneously. By incorporating prior knowledge, our method can overcome the drawbacks of sparse PCA and capture some gene modules with better biological interpretations. We evaluated the performance of ESPCA using a set of artificial datasets and two real biological datasets (including TCGA pan-cancer expression data and ENCODE expression data), and compared their performance with PCA and sparse PCA. The results showed that ESPCA could identify more biologically relevant genes, improve their biological interpretations and reveal distinct sample characteristics. An R package of ESPCA is available at http://page.amss.ac.cn/shihua.zhang/. Supplementary data are available at Bioinformatics online.

Sparse principal component regression for generalized linear models

Principal component regression (PCR) is a widely used two-stage procedure: principal component analysis (PCA), followed by regression in which the selected principal components are regarded as new explanatory variables in the model. Note that PCA is based only on the explanatory variables, so the principal components are not selected using the information on the response variable. We propose a one-stage procedure for PCR in the framework of generalized linear models. The basic loss function is based on a combination of the regression loss and PCA loss. An estimate of the regression parameter is obtained as the minimizer of the basic loss function with a sparse penalty. We call the proposed method sparse principal component regression for generalized linear models (SPCR-glm). Taking the two loss function into consideration simultaneously, SPCR-glm enables us to obtain sparse principal component loadings that are related to a response variable. However, a combination of loss functions may cause a parameter identification problem, but this potential problem is avoided by virtue of the sparse penalty. Thus, the sparse penalty plays two roles in this method. We apply SPCR-glm to two real datasets, doctor visits data and mouse consomic strain data.

Assessing the impacts of seasonal and vertical atmospheric conditions on air quality over the Pearl River Delta region

Air pollution is an increasingly concerning problem in many metropolitan areas due to its adverse public health and environmental impacts. Vertical atmospheric conditions have strong effects on vertical mixing of air pollutants, which directly affects surface air quality. The characteristics and magnitude of how vertical atmospheric conditions affect surface air quality, which are critical to future air quality projections, have not yet been fully understood. This study aims to enhance understanding of the annual and seasonal sensitivities of air pollution to both surface and vertical atmospheric conditions. Based on both surface and vertical meteorological characteristics provided by 1994–2003 monthly dynamic downscaling data from the Weather and Research Forecast Model, we develop generalized linear models (GLMs) to study the relationships between surface air pollutants (ozone, respirable suspended particulates, and sulfur dioxide) and atmospheric conditions in the Pearl River Delta (PRD) region. Applying Principal Component Regression (PCR) to address multi-collinearity, we study the contributions of various meteorological variables to pollutants’ concentration levels based on the loading and model coefficient of major principal components. Our results show that relatively high pollutant concentration occurs under relatively low mid-level troposphere temperature gradients, low relative humidity, weak southerly wind (or strong northerly wind) and weak westerly wind (or strong easterly wind). Moreover, the correlations vary among pollutant species, seasons, and meteorological variables at various altitudes. In general, pollutant sensitivity to meteorological variables is found to be greater in winter than in other seasons, and the sensitivity of ozone to meteorology differs from that of the other two pollutants. Applying our GLMs to anomalous air pollution episodes, we find that meteorological variables up to mid troposphere (∼700 mb) play an important role in influencing surface air quality, pinpointing the significant and unique associations between meteorological variables at higher altitudes and surface air quality.

Label-free monitoring of inflammatory tissue conditions using a carrageenan-induced acute inflammation rat model.

Although the confirmation of inflammatory changes within tissues at the onset of various diseases is critical for the early detection of disease and selection of appropriate treatment, most therapies are based on complex and time-consuming diagnostic procedures. Raman spectroscopy has the ability to provide non-invasive, real-time, chemical bonding analysis through the inelastic scattering of photons. In this study, we evaluate the feasibility of Raman spectroscopy as a new, easy, fast, and accurate diagnostic method to support diagnostic decisions. The molecular changes in carrageenan-induced acute inflammation rat tissues were assessed by Raman spectroscopy. Volumes of 0 (control), 100, 150, and 200 µL of 1% carrageenan were administered to rat hind paws to control the degree of inflammation. The prominent peaks at [1,062, 1,131] cm-1 and [2,847, 2,881] cm-1 were selected as characteristic measurements corresponding to the C-C stretching vibrational modes and the symmetric and asymmetric C-H (CH2 ) stretching vibrational modes, respectively. Principal component analysis of the inflammatory Raman spectra enabled graphical representation of the degree of inflammation through principal component loading profiles of inflammatory tissues on a two-dimensional plot. Therefore, Raman spectroscopy with multivariate statistical analysis represents a promising method for detecting biomolecular responses based on different types of inflammatory tissues.

Application of hyperspectral imaging and chemometrics for variety classification of maize seeds.

Seed variety classification is important for assessing variety purity and increasing crop yield. A hyperspectral imaging system covering the spectral range of 874–1734 nm was applied for variety classification of maize seeds. A total of 12 900 maize seeds including 3 different varieties were evaluated. Spectral data of 975.01–1645.82 nm were extracted and preprocessed. Discriminant models were developed using a radial basis function neural network (RBFNN). The influence of calibration sample size on classification accuracy was studied. Results showed that with the expansion of calibration sample size, calibration accuracy varied slightly, but prediction accuracy changed from the increasing form to the stable form. Accordingly, the optimal size of the calibration set was determined. Optimal wavelength selection was conducted by loading of principal components (PCs). The RBFNN model developed on optimal wavelengths with the optimal size of the calibration set obtained satisfactory results, with calibration accuracy of 93.85% and prediction accuracy of 91.00%. Visualization of classification map of seed varieties was achieved by applying this RBFNN model on the average spectra of each sample. Besides, the procedure to determine the optimal sample quantity proposed in this study was verified by support vector machine (SVM). The overall results indicated that hyperspectral imaging was a potential technique for variety classification of maize seeds, and would help to develop a real-time detection system for maize seeds as well as other crop seeds.

PD25 Principal Component Approximation: Medical Expenditure Panel Survey

Introduction:Principal component analysis (PCA) is important to summarize data or reduce dimensionality. However, one disadvantage of using PCA is the interpretability of the principal components (PCs), especially in a high-dimensional database. This study aims to analyze the patterns of variance accumulation according to PCA loadings and to approximate PCs with input variables from sample data sets.Methods:There were three data sets of various sizes used to understand the performance of PC approximation: Hitters; SF-12v2 subset of the 2004 to 2011 Medical Expenditure Panel Survey (MEPS); and, the full set of 1996 to 2011 MEPS data. The variables in three data sets were first centered and scaled before PCA. PCs approximation was studied with two approaches. First, the PC loadings were squared to estimate the variance contribution by variables to PCs. The other method was to use forward-stepwise regression to approximate PCs with all input variables.Results:The first few PCs represented large portions of total variances in each data set. Approximating PCs using stepwise regression could more efficiently identify the input variables that explain large portions of PC variances than approximating according to PCA loadings in three data sets. It required few numbers of variables to explain more than eighty percent of the PC variances.Conclusions:Approximating and interpreting PCs with stepwise regression is highly feasible. Approximating PCs can help i) interpret PCs with input variables, ii) understand the major sources of variances in data sets, iii) select unique sources of information and iv) search and rank input variables according to the proportions of PC variance explained. This is an approach to systematically understand databases and search for variables that are highly representative of databases.

Application of Near-Infrared Spectroscopy for the Detection of Metanil Yellow in Turmeric Powder

Turmeric (Curcumina Longa) is a globally traded commodity which is subjected to economically motivated chemically unsafe adulteration, namely metanil yellow. In this work, we report a simplistic and convenient approach to find the adulteration of turmeric with metanil yellow by near-infrared (NIR) spectroscopy coupled with chemometrics. Pure turmeric sample was prepared in the laboratory and spiked with different concentrations of metanil yellow. The reflectance spectra of 248 pure turmeric, metanil yellow, and adulterated samples (1–25%) (w/w) were collected using NIR spectroscopy. The calibration models based on NIR spectra of 144 samples were built for two different regression models, principal component analysis (PCR), and partial least square (PLSR) methods. Another 72 samples were used for external validation. The coefficient of determination (R 2) and root mean square error of calibration for validation and prediction were found to be 0.96–0.99, 0.44–0.91, respectively, for most of the results depending upon different pre-processing techniques and mathematical models used. The original reflectance spectra, the 1st derivative plot, the plot of PLSR regression coefficient (β), and the first three principal component loadings revealed metanil-related absorption regions. To verify the robustness of the models, the figures of merit (FOM) of the models were calculated with the help of net analyte signal (NAS) theory. Overall, it was found that PLSR yielded superior results as compared to the PCR technique. These methods can be applied to other spices also to detect the adulteration rapidly and without any prior sample preparations and with low cost.

Energy determines broad pattern of plant distribution in Western Himalaya

Several factors describe the broad pattern of diversity in plant species distribution. We explore these determinants of species richness in Western Himalayas using high‐resolution species data available for the area to energy, water, physiography and anthropogenic disturbance. The floral data involves 1279 species from 1178 spatial locations and 738 sample plots of a national database. We evaluated their correlation with 8‐environmental variables, selected on the basis of correlation coefficients and principal component loadings, using both linear (structural equation model) and nonlinear (generalised additive model) techniques. There were 645 genera and 176 families including 815 herbs, 213 shrubs, 190 trees, and 61 lianas. The nonlinear model explained the maximum deviance of 67.4% and showed the dominant contribution of climate on species richness with a 59% share. Energy variables (potential evapotranspiration and temperature seasonality) explained the deviance better than did water variables (aridity index and precipitation of the driest quarter). Temperature seasonality had the maximum impact on the species richness. The structural equation model confirmed the results of the nonlinear model but less efficiently. The mutual influences of the climatic variables were found to affect the predictions of the model significantly. To our knowledge, the 67.4% deviance found in the species richness pattern is one of the highest values reported in mountain studies. Broadly, climate described by water–energy dynamics provides the best explanation for the species richness pattern. Both modeling approaches supported the same conclusion that energy is the best predictor of species richness. The dry and cold conditions of the region account for the dominant contribution of energy on species richness.

Pregnancy probabilistically augments potential precursors to chronic, immune-mediated or autoimmune lacrimal gland infiltrates

PurposeThis study asked whether pregnancy, a risk factor for dry eye disease associated with both chronic, immune-mediated- and autoimmune etiologies, augments development of clusters of coordinately functioning cells (CCFC) that may be precursors to pathological lacrimal gland infiltrates. MethodsLacrimal glands were from six virgin- and six term-pregnant rabbits of the same age and environmental exposure history. Seventy-two immune response-related gene transcripts were assayed by real time RT-PCR. Principal component (PC) analysis identified transcript signatures of CCFC contributing negative (⊖) or positive (⊕) PC loadings and determined gland PC projections, which reflect levels of CCFC development. ResultsThree CCFC were of interest as potential precursors to pathological infiltrates. CCFC 1⊖ was suggestive of an ectopic lymphoid structure with resting T cells and B cells. CCFC 1⊕ was suggestive of an immune-mediated infiltrate with TH1 cells and mature, cytotoxic B cells. CCFC 2⊖ was suggestive of an ectopic lymphoid structure with activated T cells, mature B cells, germinal center, and plasmacytes. CCFC 4⊖ and CCFC 5⊖ also included plasmacytes. Pregnancy augmented CCFC 1⊖ in some glands; augmented CCFC 1⊕ in others; and augmented CCFC 2⊖, CCFC 4⊖, and CCFC 5⊖ different combinations. ConclusionsPotential precursors of pathological infiltrates form in the lacrimal glands by the time of sexual maturity. Pregnancy augments lacrimal gland plasmacyte populations, and it can augment development of potential precursors to either chronic, immune-mediated infiltrates or autoimmune infiltrates of various phenotypes. Systemic and strictly local, probabilistic phenomena interact with pregnancy to determine which combinatorial phenotypes are favored.

USE AND MULTIFACTORIAL RECONCILIATION OF UNIFACE REDUCTION MEASURES: A PILOT STUDY AT THE NOBLES POND PALEOINDIAN SITE

How much stone tools are reduced and their form changed from first use to discard bears upon issues such as typological integrity, curation rate, and effects of occupation span. But degree of reduction depends partly upon the measures used to gauge it. Most studies involve single indices that gauge reduction in different ways or at different scales, so results are difficult to compare between studies. In this pilot study, we compare four allometric reduction measures—one each based on length, length:thickness ratio, volume, and mass, estimated by comparing observed values in discarded tools to estimated original values—for consistency when applied to an endscraper sample from the Nobles Pond Paleoindian site in Ohio, USA. Fitted to the Weibull distribution, all measures suggest attrition compared to experimental controls, but variation among them demands reconciliation. A multifactorial method that weights individual measures by their principal-component loadings suggests attritional discard at increasing rate as reduction advances. More importantly, it addresses the growing problem of reconciling the many reduction measures in use, a major concern in this expanding research area.

Incorporating biological information in sparse principal component analysis with application to genomic data

BackgroundSparse principal component analysis (PCA) is a popular tool for dimensionality reduction, pattern recognition, and visualization of high dimensional data. It has been recognized that complex biological mechanisms occur through concerted relationships of multiple genes working in networks that are often represented by graphs. Recent work has shown that incorporating such biological information improves feature selection and prediction performance in regression analysis, but there has been limited work on extending this approach to PCA. In this article, we propose two new sparse PCA methods called Fused and Grouped sparse PCA that enable incorporation of prior biological information in variable selection.ResultsOur simulation studies suggest that, compared to existing sparse PCA methods, the proposed methods achieve higher sensitivity and specificity when the graph structure is correctly specified, and are fairly robust to misspecified graph structures. Application to a glioblastoma gene expression dataset identified pathways that are suggested in the literature to be related with glioblastoma.ConclusionsThe proposed sparse PCA methods Fused and Grouped sparse PCA can effectively incorporate prior biological information in variable selection, leading to improved feature selection and more interpretable principal component loadings and potentially providing insights on molecular underpinnings of complex diseases.

Land eco-security evaluation in the Shule River Basin during 2005–2014 in Gansu Province, China

ABSTRACTThe Shule River Basin is an ecologically fragile area in arid zone. To understand the state of land eco-security, the Environment-Economic-Society model was applied to build a land eco-security evaluation index system in the Shule River Basin. An entropy-weighted and matter-element model was built for eco-security evaluation from 2005 to 2014. Principal component analysis was used to quantitatively study the limiting factors of land ecological security. The result showed: the direction of development of the state of land eco-security in the Shule River Basin from 2005 to 2014 was characterized by “unsafe (No4) → safe (No1),” and presented an upward trend. The land eco-security status during 2005–2007 was in the “Unsafe” state and the state changed to “Critical Safe” in 2008–2009, “Safer” in 2010–2011, but “Safe” in 2012–2014. The key factors that affected land eco-security in the Shule River Basin were Per Capita Arable Land, Forest Cover Rate, Per Capita Water Resources, Water Production Modulus, the Tertiary Industry Output Value, and GDP Ratio and Water Consumption. Among them, Forest Cover Rate and Water Production Modulus had the greatest impact, with principal component loads of up to 0.973 and 0.968, respectively. The result of this study is expected to serve as reference and support for the conservation and management of Shule River Basin to ensure sustainable development.

Selection of minimum indicators of hydrologic alteration of the Gorai river, Bangladesh using principal component analysis

The use of a few indicators of hydrologic alteration can minimize statistical redundancy which can lead to take effective environmental flow management decisions. The aim of the research is to select representative dominant indicators of hydrological alteration through reducing the dimensionality of multivariate data for the Gorai river, Bangladesh. The required mean daily discharge data (1976–2006) of the Gorai have been collected from the Bangladesh Water Development Board. The Range of Variability Approach and anomaly of the river flow show hydrologic alteration in recent years comparing with past. Principal component analysis (PCA) in Kendall and Pearson methods has been performed by the XLSTAT software (version 2.9). Depending on eigenvalues (>1.0), seven and six Principal components (PCs) are selected for Kendall and Pearson methods, respectively, for further analysis. Since the eco-flow metrics of eco-deficit of the river has poor correlation with the indicators of hydrologic alteration, the resultant loadings are subjected to the Kaiser–Meyer–Olkin (KMO) Measure of Sampling Adequacy (MSA) method which is classified into marvelous, meritorious, middling, mediocre, and miserable based on the PC loading values. Seven indicators have been found whose PC loadings are in marvelous class and are common for both Kendall and Pearson methods. These indicators, named March mean flow, 1-day minimum mean, 3-day minimum mean, 7-day minimum mean, 30-day minimum mean, 90-day minimum mean, and 30-day maximum mean flows, are the desired indicators of hydrologic alteration of the Gorai river.

HYPERSPECTRAL CLASSIFICATION FOR IDENTIFYING DECAYED ORANGES INFECTED BY FUNGI

Fast and nondestructive detection of early decay caused by fungal infection in citrus fruit was a challenging task for the citrus industry during the postharvest fruit processing. In general, workers relied on the ultraviolet induction fluorescence technique to detect and remove the decayed citrus fruits in fruit packing houses. However, this operation was harmful for human health, and was also very inefficient. In this study, navel oranges were used as research object. A novel method combining with hyperspectral imaging technology in the wavelength region between 400 and 1100 nm wavelength was proposed to solve this problem. First, normalization approaches were applied to decrease the variation of spectral reflectance intensity due to natural curvature of navel orange surface. Then, the spectral data of regions of interest (ROIs) from normal and decayed tissues was analyzed by principal component analysis (PCA) for investigating the performance of visible and near infrared (Vis-NIR) hyperspectral data to discriminate these two kinds of tissues. Next, six characteristic wavelength images were obtained by analyzing the loadings of the first principal component (PC1). And, a multispectral image was established by using the corrected six characteristic wavelength images. On basis of the multispectral image, pseudo-color image processing with intensity slicing was utilized to produce a two-dimensional color image with clear contrast between decayed and normal tissues. Finally, an image segmentation algorithm by combining the pseudo-color processing method and a global threshold method was proposed for fast identification of decayed navel oranges. For 240 independent samples, the success rates were 100 and 97.5% for decayed navel oranges infected by Penicillium digitatum and normal navel oranges, respectively. In particular, the proposed algorithm was also applied to detect the decayed navel oranges infected by Penicillium italicum (samples not used for the development of algorithm) and obtained a 91.7% identification accuracy, indicating a well generalization ability and actual application value of the proposed algorithm.