Standard Principal Component Analysis Research Articles

Compressed representation of brain genetic transcription.

The architecture of the brain is too complex to be intuitively surveyable without the use of compressed representations that project its variation into a compact, navigable space. The task is especially challenging with high-dimensional data, such as gene expression, where the joint complexity of anatomical and transcriptional patterns demands maximum compression. The established practice is to use standard principal component analysis (PCA), whose computational felicity is offset by limited expressivity, especially at great compression ratios. Employing whole-brain, voxel-wise Allen Brain Atlas transcription data, here we systematically compare compressed representations based on the most widely supported linear and non-linear methods-PCA, kernel PCA, non-negative matrix factorisation (NMF), t-stochastic neighbour embedding (t-SNE), uniform manifold approximation and projection (UMAP), and deep auto-encoding-quantifying reconstruction fidelity, anatomical coherence, and predictive utility across signalling, microstructural, and metabolic targets, drawn from large-scale open-source MRI and PET data. We show that deep auto-encoders yield superior representations across all metrics of performance and target domains, supporting their use as the reference standard for representing transcription patterns in the human brain.

Human and ecological risk assessments of potentially toxic elements in sediments around a pharmaceutical industry

Potentially toxic elements (PTEs) in sediment can be highly hazardous to the environment and public health. This study aimed to assess the human and ecological risks of PTEs in sediments around a pharmaceutical industry in Ilorin, Nigeria. Physicochemical parameters and the concentrations of lead (Pb), chromium (Cr), cadmium (Cd), cobalt (Co), arsenic (As), and nickel (Ni) were analyzed in sediment samples collected from seven locations in the wet and dry seasons. Standard two-dimensional principal component analysis (PCA) and risk assessments were also conducted. The concentrations of Pb, Co, Ni, Cr, Cd, and As in the sediments ranged from 0.001 to 0.031 mg/kg, 0–0.005 mg/kg, 0.005–0.012 mg/kg, 0.001–0.014 mg/kg, 0.005–0.024 mg/kg, and 0.001–0.012 mg/kg, respectively. The mean concentrations of the total PTEs content were found in decreasing order of concentration: Pb > Cd > Ni > Cr > As > Co. PCA showed that some of the PTEs were highly concentrated in samples obtained at other locations as well as at the discharge point. The Hazard Index was mostly <1 across locations, indicating little to no probable non-cancerous effect. However, the incremental lifetime cancer risk for arsenic and nickel was high and required attention. The ecological risk assessment showed that lead and arsenic were the major PTEs pollutants in all locations. The study identifies PTEs profiles in sediments and emphasises the necessity of continual monitoring and action to stop long-term negative impacts on the local environment and public health.

Получение и обработка акустических откликов волн Лэмба в датчиках водных растворов базовых вкусов

In this work, the total acoustic wave losses in 128°Y-LiNbO3 wafers during liquid application due to radiation emission of waves into the liquid medium, as well as its viscosity, conductivity, density and dielectric constant of the liquid analyte are measured. Solutions with 5 basic flavors, widely used for calibration and reference creation in liquid media testing, were selected as test liquids. The measurements were performed in the range of 30...60 MHz. After the measurements, the experimental data were presented in the form of polar histograms, from the comparison of areas and shapes of which the flavor of a particular liquid was determined. Then, a principal component analysis (PCA) method was applied to the same experimental results to describe linear elastic wave deformation processes based on experimental data containing known values of physical wave measurements for different types of liquid media, which was previously used to identify gases and predict the properties of various substances from the linear response of a normal acoustic-electronic wave. In this paper, a comparative analysis of the results obtained using polar histograms and the standard PCA classification method is performed. The advantages of the latter are demonstrated. The results of this study are useful both for the development of acoustic sensors without sensor layers and for the application of machine learning methods to data in other types of sensors.

Machine learning framework for the real-time reconstruction of regional 4D ocean temperature fields from historical reanalysis data and real-time satellite and buoy surface measurements

The ability to model 4D properties of near-coast ocean regions in real time is valuable for both general environmental monitoring and for specialized industry applications. We present a computationally parsimonious framework to build improved spatiotemporal models that combine dynamics inferred from historical high-fidelity model outputs and real-time sensor measurements. We are motivated by a temperature data set of the Massachusetts Bay from sensors that can only obtain measurements at the ocean’s surface. We first apply standard principal component analysis (PCA) to a historical reanalysis data set of the time-evolving 3D (x,y,z) temperature field. Next, we train a temporal convolutional neural network (TCN) to predict the time coefficients of the vertical PCA modes, and their variance, as a function of surface temperature. Simultaneously, we estimate the time-evolving 2D (x,y) surface temperature field from the satellite and buoy measurements with multi-fidelity Gaussian process regression (GPR). Finally, the surface temperature is used as input to the neural network to probabilistically predict the PCA coefficients and reconstruct the full 4D temperature field. The results are compared to in-situ measurements at all depths, and the median absolute error is found to be 0.97°C. Overall, the proposed framework is able to produce more accurate models in less time, and the uncertainty quantification can be leveraged for future decision-making.

Tensorial Principal Component Analysis in Detecting Temporal Trajectories of Purchase Patterns in Loyalty Card Data: Retrospective Cohort Study.

Loyalty card data automatically collected by retailers provide an excellent source for evaluating health-related purchase behavior of customers. The data comprise information on every grocery purchase, including expenditures on product groups and the time of purchase for each customer. Such data where customers have an expenditure value for every product group for each time can be formulated as 3D tensorial data. This study aimed to use the modern tensorial principal component analysis (PCA) method to uncover the characteristics of health-related purchase patterns from loyalty card data. Another aim was to identify card holders with distinct purchase patterns. We also considered the interpretation, advantages, and challenges of tensorial PCA compared with standard PCA. Loyalty card program members from the largest retailer in Finland were invited to participate in this study. Our LoCard data consist of the purchases of 7251 card holders who consented to the use of their data from the year 2016. The purchases were reclassified into 55 product groups and aggregated across 52 weeks. The data were then analyzed using tensorial PCA, allowing us to effectively reduce the time and product group-wise dimensions simultaneously. The augmentation method was used for selecting the suitable number of principal components for the analysis. Using tensorial PCA, we were able to systematically search for typical food purchasing patterns across time and product groups as well as detect different purchasing behaviors across groups of card holders. For example, we identified customers who purchased large amounts of meat products and separated them further into groups based on time profiles, that is, customers whose purchases of meat remained stable, increased, or decreased throughout the year or varied between seasons of the year. Using tensorial PCA, we can effectively examine customers' purchasing behavior in more detail than with traditional methods because it can handle time and product group dimensions simultaneously. When interpreting the results, both time and product dimensions must be considered. In further analyses, these time and product groups can be directly associated with additional consumer characteristics such as socioeconomic and demographic predictors of dietary patterns. In addition, they can be linked to external factors that impact grocery purchases such as inflation and unexpected pandemics. This enables us to identify what types of people have specific purchasing patterns, which can help in the development of ways in which consumers can be steered toward making healthier food choices.

Measuring systemic risk contribution: A higher-order moment augmented approach

Marginal contributions of individual institutions to the systemic risk contain predictive power for potential future exposures and provide early warning signals to regulators and the public. In this paper, the higher-order co-skewness and co-kurtosis are used to construct systemic risk contribution measures, which allow us to identify and characterize the co-movement driving the asymmetry and tail behavior of the joint distribution of asset returns. We illustrate the usefulness of higher-order moment augmented approach by using 4868 stocks living in the Chinese market from June 2002 to March 2022. The empirical results show that these higher-order moment measures convey useful information for systemic risk contribution measurement and portfolio selection, complementary to the information extracted from a standard principal components analysis.

Fault Detection and Diagnosis in Tennessee Eastman Process with Deep Autoencoder

Data-driven modeling has been considered as an attractive approach for fault detection in chemical processes. Of special interest to industry are methods that represent nonlinear phenomena and detect complex faults. In this paper, a semi-supervised deep learning method - deep autoencoder for fault detection in Tennessee Eastman Process (TEP) is proposed. The TEP process is a simulated benchmark for evaluating process control and monitoring methods. The performance of the proposed method is evaluated and compared to Principal Component Analysis (PCA). The experimental results demonstrate that the proposed optimized five-layers DAE model for fault detection outperforms the standard PCA. Of special importance to real-world applications is its capability for automatic variable selection. In comparison to PCA it demonstrated higher prediction accuracy for most of the generated faults. Deep autoencoder has the potential to become an excellent approach for process monitoring and fault detection in chemical processes.

CONCERTO: Extracting the power spectrum of the [CII] emission line

Context. CONCERTO is the first experiment to perform a [CII] line intensity mapping (LIM) survey on the COSMOS field to target z &gt; 5.2. Measuring the [CII] angular power spectrum allows us to study the role of dusty star-forming galaxies in the star formation history during the epochs of Reionization and post-Reionization. The main obstacle to this measurement is the contamination by bright foregrounds: the dust continuum emission and atomic and molecular lines from foreground galaxies at z ≲ 3. Aims. We evaluate our ability to retrieve the [CII] signal in mock observations of the sky using the Simulated Infrared Dusty Extragalactic Sky (SIDES), which covers the mid-infrared to millimetre range. We also measure the impact of field-to-field variance on the residual foreground contamination. Methods. We compared two methods for dealing with the dust continuum emission from galaxies (i.e. the cosmic infrared background fluctuations): the standard principal component analysis (PCA) and the asymmetric re-weighted penalized least-squares (arPLS) method. For line interlopers, the strategy relies on masking low-redshift galaxies using the instrumental beam profile and external catalogues. As we do not have observations of CO or deep-enough classical CO proxies (such as LIR), we relied on the COSMOS stellar mass catalogue, which we demonstrate to be a reliable CO proxy for masking. To measure the angular power spectrum of masked data, we adapted the P of K EstimatoR (POKER) from cosmic infrared background studies and discuss its use on LIM data. Results. The arPLS method achieves a reduction in the cosmic infrared background fluctuations to a sub-dominant level of the [CII] power at z ∼ 7, a factor of &gt; 70 below our fiducial [CII] model. When using the standard PCA, this factor is only 0.7 at this redshift. The masking lowers the power amplitude of line contamination down to 2 × 10−2 Jy2 sr−1. This residual level is dominated by faint undetected sources that are not clustered around the detected (and masked) sources. For our [CII] model, this results in a detection at z = 5.2 with a power ratio [CII]/(residual interlopers) = 62 ± 32 for a 22% area survey loss. However, at z = 7, [CII]/(residual interlopers) = 2.0 ± 1.4, due to the weak contrast between [CII] and the residual line contamination. Thanks to the large area covered by SIDES-Uchuu, we show that the power amplitude of line residuals varies by 12–15% for z = 5.2 − 7, which is less than the field-to-field variance affecting [CII] power spectra. Conclusions. We present an end-to-end simulation of the extragalactic foreground removal that we ran to detect the [CII] at high redshift via its angular power spectrum. We show that cosmic infrared background fluctuations are not a limiting foreground for [CII] LIM. On the contrary, the CO and [CI] line contamination severely limits our ability to accurately measure the [CII] angular power spectrum at z ≳ 7.

Fundamental limits in structured principal component analysis and how to reach them

How do statistical dependencies in measurement noise influence high-dimensional inference? To answer this, we study the paradigmatic spiked matrix model of principal components analysis (PCA), where a rank-one matrix is corrupted by additive noise. We go beyond the usual independence assumption on the noise entries, by drawing the noise from a low-order polynomial orthogonal matrix ensemble. The resulting noise correlations make the setting relevant for applications but analytically challenging. We provide characterization of the Bayes optimal limits of inference in this model. If the spike is rotation invariant, we show that standard spectral PCA is optimal. However, for more general priors, both PCA and the existing approximate message-passing algorithm (AMP) fall short of achieving the information-theoretic limits, which we compute using the replica method from statistical physics. We thus propose an AMP, inspired by the theory of adaptive Thouless-Anderson-Palmer equations, which is empirically observed to saturate the conjectured theoretical limit. This AMP comes with a rigorous state evolution analysis tracking its performance. Although we focus on specific noise distributions, our methodology can be generalized to a wide class of trace matrix ensembles at the cost of more involved expressions. Finally, despite the seemingly strong assumption of rotation-invariant noise, our theory empirically predicts algorithmic performance on real data, pointing at strong universality properties.

The attributes senior high school mathematics teachers and their students value in mathematics learning

Valuing in mathematics is one of the areas in mathematics education that has attracted the attention of mathematics education researchers in recent times. While efforts have been made to research the attributes students value in their mathematics learning, the same cannot be said about attributes teachers and their students value in mathematics learning, and value alignment in mathematics learning between teachers and students. This paper reports on a study that sought to explore what teachers and their students value and the alignment between attributes teachers and their students value in learners’ mathematics learning. A survey of what students and their teachers value in learners’ mathematics learning was conducted with 690 senior high school students and 66 teachers from nine schools, using adapted What I Find Important in my mathematics learning questionnaires. A multistage sampling procedure involving the use of stratified random sampling and purposive sampling procedures was used to select the research participants. The data obtained from the study were then analysed, using frequency counts, mean and standard deviation and principal component analysis. The results from the study revealed that the senior high school mathematics teachers valued Authority, Engagement, Strategy and Communication in students’ mathematics learning, while the students valued Authority, Relevance, Understanding, Engagement, Strategies, ICT and Achievement in their own mathematics learning; an indication of a gap in valuing among the two groups of research participants. Implications of the findings for mathematics teaching and learning as well as in-service and pre-service education programmes are provided.

Principal component analysis in the epidemiology of diarrhoea in calves

The paper is the first to use principal component analysis (PCA) in veterinary epidemiology and aims to identify the most significant agents of diarrhoea in calves. Data were collected from 245 calves on 32 farms within 13 districts of Northern Kazakhstan. Epidemiological data were simulated in R, using standard PCA modules. The data clustering was carried out based on the prevalence of seven enteropathogens for an age group dataset (represented by four animal groups) and a farm type dataset (by three farm types depending on herd size). The simulation revealed that the components identified here for one and two-week calves explained 91.31% of the variance. The first two components of large and middle-sized farms covered 90.3% of the variance. The coordinates corresponding to pathogens were approximately visualised in directions of eigenvectors for each age group and farm type. The coordinate for Cryptosporidium parvum was in directions of eigenvectors for one-to three-week calves and large farms. The coordinate for rotaviruses was in the direction of eigenvectors for four-week calves and medium-sized and small farms. So, PCA can potentially be useful in the clustering of epidemiological datasets and making decisions on control of infectious diseases with a multi-pathogenic nature.

Sparse principal component analysis for high‐dimensional stationary time series

AbstractWe consider the sparse principal component analysis for high‐dimensional stationary processes. The standard principal component analysis performs poorly when the dimension of the process is large. We establish oracle inequalities for penalized principal component estimators for the large class of processes including heavy‐tailed time series. The rate of convergence of the estimators is established. We also elucidate the theoretical rate for choosing the tuning parameter in penalized estimators. The performance of the sparse principal component analysis is demonstrated by numerical simulations. The utility of the sparse principal component analysis for time series data is exemplified by the application to average temperature data.

Farm Household Typology Based on Soil Quality and Influenced by Socio-Economic Characteristics and Fertility Management Practices in Eastern Kenya

The smallholder farming systems in Sub-Saharan Africa (SSA) are highly diverse and heterogeneous in terms of biophysical and socio-economic characteristics. This study was conducted in upper Eastern Kenya (UEK) to categorize farm households and determine the influence of socio-economic characteristics (SeC) and soil fertility management practices (SFMP) on soil fertility across farms. Conditioned Latin hypercube sampling (cLHS) was performed to determine 69 soil sampling sites within Meru and Tharaka Nithi counties. From each household (whose field soil sample was obtained), data relating to resource endowment and soil fertility management were collected through a household questionnaire survey. Standard laboratory procedures were used to analyse soil samples. Data reduction was performed using categorical principal component analysis (CATPCA) (for SeC and SFMP) and standard principal component analysis (PCA) (for soil properties). Two-step cluster analysis identified three distinct farm categories or farm types (FT), namely, low fertility farms (FT1), moderately fertile farms (FT2), and fertile farms (FT3). The correlation of clusters against soil properties was significant across pH, soil organic carbon (SOC), cation exchange capacity (CEC), available P, plant available K, and exchangeable bases. FT1 had low SOC, pH, CEC and available P (soil characteristics), low usage of fertilizer and manure (soil fertility management), and smaller household size, lower income, and smaller farm size (socio-economic). FT2 had lower SOC (compared to FT3) and available P. In terms of soil fertility management, FT2 had higher cases of fallowing and composting with moderate fertilizer usage. Households in this category had moderate income, family size, and land size (socio-economic). FT3 had relatively high SOC, pH, CEC, and mineral nutrients. This farm type was characterized by high fertilizer use (soil fertility management) as well as larger household size, higher income, and larger farm size (socio-economic). The results indicate the importance of nutrient management in enhancing soil quality. Delineation and characterization of farms based on the various parameters including resource endowment reveal imbalanced farm resource flows, suggesting a need for locally tailored interventions suited for location-specific conditions to facilitate improved targeting of soil fertility-enhancing technologies and sustainable crop production regimes. While fertilizer is one of the most critical inputs for enhancing agricultural production, it is a major contributor to nitrous oxide emissions from agriculture and can have negative environmental effects on soil biota and water sources. Farmers’ knowledge on the use of fertilizer is thus necessary in developing strategies (such as integrated approach) to promote its efficient use and minimize its detrimental impacts.

Penalized optimal scaling for ordinal variables with an application to international classification of functioning core sets.

Ordinal data occur frequently in the social sciences. When applying principal component analysis (PCA), however, those data are often treated as numeric, implying linear relationships between the variables at hand; alternatively, non-linear PCA is applied where the obtained quantifications are sometimes hard to interpret. Non-linear PCA for categorical data, also called optimal scoring/scaling, constructs new variables by assigning numerical values to categories such that the proportion of variance in those new variables that is explained by a predefined number of principal components (PCs) is maximized. We propose a penalized version of non-linear PCA for ordinal variables that is a smoothed intermediate between standard PCA on category labels and non-linear PCA as used so far. The new approach is by no means limited to monotonic effects and offers both better interpretability of the non-linear transformation of the category labels and better performance on validation data than unpenalized non-linear PCA and/or standard linear PCA. In particular, an application of penalized optimal scaling to ordinal data as given with the International Classification of Functioning, Disability and Health (ICF) is provided.

Genetic relationship and diversity of cultivars and breeding lines of tetraploid Italian ryegrass (Lolium multiflorum Lam.) and its hybrids with Lolium‐Festuca complex based on genome‐wide allele frequency

AbstractGenetic relatedness and diversity of 62 cultivars and breeding lines of tetraploid Italian ryegrass (Lolium multiflorum Lam.; 39 accessions) and its interspecific hybrids, Festulolium (18 accessions), and hybrid ryegrass (Lolium × hybridum Hausskn.; 5 accessions), mainly from Japan, were revealed based on 2,824 genome‐wide allele frequencies obtained by the genotyping by random amplicon sequencing‐direct (GRAS‐Di) method using bulk genomic DNA testing. Genomic composition of each accession was estimated according to the occurrence of 77,373 unique GRAS‐Di sequences in the reference population consisting of diploid Italian ryegrass, meadow fescue (Festuca pratensis Huds.), and perennial ryegrass (Lolium perenne L.). The high correlation coefficient (0.98) between the fescue‐specific reads ratio and the previously obtained f‐ratio of genomic in situ hybridization in Festulolium cultivars suggests the usefulness of this simple method. Both cluster analysis based on Nei's standard genetic distance (DST) and principal component analysis (PCA) showed that groups were formed largely by species. However, the complex heritage of Lolium‐Festuca (Festulolium) materials could not be determined by species registration or breeding history alone. Some Festulolium accessions were closely related to Italian ryegrass, whereas some defined as Italian ryegrass may actually be interspecific hybrids. The high genetic diversity of Festulolium compared to Italian ryegrass and hybrid ryegrass revealed by PCA seems due to the wide range of fescue‐specific read ratios (0.04–33.0%). Tetraploid Italian ryegrass did not show clear structural differentiation, but some genetic relationships attributable to breeding history were demonstrated. Mean pairwise DST of tetraploid Italian ryegrass cultivars was significantly lower than that of diploids. Tetraploids and diploids could be separated by PCA plot. Although mean expected heterozygosities of tetraploid and diploid cultivars were not significantly different, the results suggest that the utilization of diploid genetic resources is effective in maintaining and increasing the genetic diversity of breeding populations of tetraploid Italian ryegrass.

End-point determination of the extraction processes for Stevia rebaudiana Bertoni leaves by near-infrared spectroscopy

Rebaudioside A (RA) and Stevioside (STV) are abundant steviol glycosides (SGs) contained in Stevia rebaudiana Bertoni leaves, which exhibit good stability and various pharmacological activities that have been widely developed in the food and pharmaceutical industries. However, the stevia industry still suffers from high consumption, low efficiency, and long-term dependence on the operational experience of workers. The extraction process of Stevia rebaudiana Bertoni leaves one of the fundamental units for the production of SGs, which is crucial for the homogeneous stability of the final product. The applications of near-infrared (NIR) spectroscopy combined with chemometrics in determining the end-point of the extraction process were proposed in this study. Firstly, the quantitative models were established by partial least squares regression (PLSR) and back-propagation artificial neural network (BP-ANN) to rapidly detect the changes of content RA and STV, which have good prediction results. Secondly, the qualitative analysis methods were established by moving block of standard deviation (MBSD), absolute distance of standard deviation (ADSD) and principal component analysis (PCA) to rapidly determine the extraction end-point, which MBSD method was consistent with the high-performance liquid chromatography (HPLC) method. Finally, the variation of the extraction process was revealed by Aquaphotomic to provide a microscopic perspective of water molecules for end-point determination. The HPLC method requires 30 min to determine the content, whereas NIR spectroscopy requires only 18 s to obtain a spectrum. These results indicate that the extraction end-point of Stevia rebaudiana Bertoni leaves can be determined rapidly and accurately using NIR spectroscopy, which provides a significant reference for other food, medicinal plants, and agricultural products production processes.

The BINGO project

Aims.The large-scale distribution of neutral hydrogen (H I) in the Universe is luminous through its 21 cm emission. The goal of the Baryon Acoustic Oscillations from Integrated Neutral Gas Observations (BINGO) radio telescope is to detect baryon acoustic oscillations at radio frequencies through 21 cm intensity mapping (IM). The telescope will span the redshift range 0.127 &lt; z &lt; 0.449 with an instantaneous field-of-view of 14.75° ×6.0°.Methods.In this work we investigate different constructive and operational scenarios of the instrument by generating sky maps as they would be produced by the instrument. In doing this we use a set of end-to-end IM mission simulations. The maps will additionally be used to evaluate the efficiency of a component separation method (GNILC).Results.We have simulated the kind of data that would be produced in a single-dish IM experiment such as BINGO. According to the results obtained, we have optimized the focal plane design of the telescope. In addition, the application of theGNILCmethod on simulated data shows that it is feasible to extract the cosmological signal across a wide range of multipoles and redshifts. The results are comparable with the standard principal component analysis method.

Haplotype and population structure inference using neural networks in whole-genome sequencing data.

Accurate inference of population structure is important in many studies of population genetics. Here we present HaploNet, a method for performing dimensionality reduction and clustering of genetic data. The method is based on local clustering of phased haplotypes using neural networks from whole-genome sequencing or dense genotype data. By using Gaussian mixtures in a variational autoencoder framework, we are able to learn a low-dimensional latent space in which we cluster haplotypes along the genome in a highly scalable manner. We show that we can use haplotype clusters in the latent space to infer global population structure using haplotype information by exploiting the generative properties of our framework. Based on fitted neural networks and their latent haplotype clusters, we can perform principal component analysis and estimate ancestry proportions based on a maximum likelihood framework. Using sequencing data from simulations and closely related human populations, we show that our approach is better at distinguishing closely related populations than standard admixture and principal component analysis software. We further show that HaploNet is fast and highly scalable by applying it to genotype array data of the UK Biobank.

Improved Interpretability of Brain-Behavior CCA With Domain-Driven Dimension Reduction.

Canonical Correlation Analysis (CCA) has been widely applied to study correlations between neuroimaging data and behavioral data. Practical use of CCA typically requires dimensionality reduction with, for example, Principal Components Analysis (PCA), however, this can result in CCA components that are difficult to interpret. In this paper, we introduce a Domain-driven Dimension Reduction (DDR) method, reducing the dimensionality of the original datasets and combining human knowledge of the structure of the variables studied. We apply the method to the Human Connectome Project S1200 release and compare standard PCA across all variables with DDR applied to individual classes of variables, finding that DDR-CCA results are more stable and interpretable, allowing the contribution of each class of variable to be better understood. By carefully designing the analysis pipeline and cross-validating the results, we offer more insights into the interpretation of CCA applied to brain-behavior data.

Multi-band PCA based ear recognition technique

Principal Component Analysis (PCA) has been successfully applied to many applications, including ear recognition. This paper presents a Two Dimensional Multi-Band PCA (2D-MBPCA) method, inspired by PCA based techniques for multispectral and hyperspectral images, which have demonstrated significantly higher performance to that of standard PCA. The proposed method divides the input image into a number of images based on the intensity of the pixels. Three different methods are used to calculate the pixel intensity boundaries, called: equal size, histogram, and greedy hill climbing based techniques. Conventional PCA is then applied on the resulting images to extract their eigenvectors, which are used as features. The optimal number of bands was determined using the intersection of number of features and total eigenvector energy. Experimental results on two benchmark ear image datasets demonstrate that the proposed 2D-MBPCA technique significantly outperforms single image PCA by up to 56.41% and the eigenfaces technique by up to 29.62% with respect to matching accuracy on images from two benchmark datasets. Furthermore, it gives very competitive results to those of learning based techniques at a fraction of their computational cost and without a need for training.