Second-order Spectra Research Articles

Bearing fault detection using second-order moment spectrum and adaptive time-varying morphological filtering

The second-order moment spectrum is a method designed to simplify the complex shape of the spectrum, thus facilitating its interpretation for the identification and localization of defects based on peak frequency. Generally used as a final step in defect detection methods, this method offers the advantage of a more easily interpretable spectral shape. Compared to the shape of the spectrum of the vibration signal defined by the Fourier transform, which includes sidebands composed of peaks of large amplitude at different frequencies, the spectrum generated by the second-order moment spectrum method stands out for its simplicity. Starting from the mean and standard deviation of the vibration signal, the second-order moment can be defined as the power of the ratio between the standard deviation and the difference between the signal and the mean. Next, the Fourier transform is applied to express the second-order moment spectrum. The performance of the second-order moment spectrum is evaluated using the principle of comparison with the envelope spectrum obtained by the Hilbert transform. Vibration signals are analyzed using two methods: adaptive time-varying morphological filtering and second-order moment spectrum. After applying these methods to the signals from the database, we observe high-amplitude peaks at the frequencies corresponding to inner ring and ball defects. The second-order moment spectrum gives similar results to those obtained with the Hilbert transform envelope.

Eco-friendly second derivative synchronous fluorescence spectroscopic method for simultaneous determination of rosuvastatin and ezetimibe in pharmaceutical capsules

The fixed dose combination of Rosuvastatin and ezetimibe has recently received approval from the FDA for the treatment of elevated levels of low-density lipoprotein cholesterol in adults. Herein, an eco-friendly and highly sensitive spectrofluorimetric method was developed and validated for simultaneous determination of rosuvastatin and ezetimibe in commercial capsules. The developed method involved synchronous fluorescence spectroscopy combined with second derivative spectroscopy to resolve the overlapping fluorescence spectra of rosuvastatin and ezetimibe. The studied drugs were measured in synchronous mode at Δλ of 40 and their recorded synchronous fluorescence spectra were derivatized into second-order spectra, enabling the selective quantification of rosuvastatin and ezetimibe at 370 nm and 312 nm, respectively. Optimization studies regarding to the influence of buffer pH, incorporation of surfactant, choice of diluting solvent, and synchronous Δλ were carried out. The method was validated using the validation characteristics listed in ICH Q2(R1). The calibration curves displayed satisfactory linear relationships across the calibration range of 0.1–2 µg/mL for rosuvastatin and 0.05–3 µg/mL for ezetimibe. The methodology demonstrated robustness to minor modifications in the procedural parameters and selectivity in quantifying the studied drugs in synthetic mixed solutions and commercial capsules without interference. Furthermore, the level of environmental friendliness and sustainability of the suggested spectrofluorimetric approach was assessed in relation to two previously documented methodologies utilizing the AGREE metric. The findings indicated that the suggested method demonstrated a notably superior level of sustainability in comparison to the documented methods.

How Thick is the Air-Water Interface?─A Direct Experimental Measurement of the Decay Length of the Interfacial Structural Anisotropy.

The air-water interface is a highly prevalent phase boundary impacting many natural and artificial processes. The significance of this interface arises from the unique properties of water molecules within the interfacial region, with a crucial parameter being the thickness of its structural anisotropy, or "healing depth". This quantity has been extensively assessed by various simulations which have converged to a prediction of a remarkably short length of ∼6 Å. Despite the absence of any direct experimental measurement of this quantity, this predicted value has surprisingly become widely accepted as fact. Using an advancement in nonlinear vibrational spectroscopy, we provide the first measurement of this thickness and, indeed, find it to be ∼6-8 Å, finally confirming the prior predictions. Lastly, by combining the experimental results with depth-dependent second-order spectra calculated from ab initio parametrized molecular dynamics simulations, which are also in excellent agreement with this experimental result, we shed light on this surprisingly short correlation length of molecular orientations at the interface.

A comprehensive and innovative chemometric approach: Archaeometric analysis of the sherds from the Neolithic Period to the Chalcolithic and early Bronze Age with the full deployment of FTIR's molecular spectroscopic capabilities

In this study, the mineralogical composition of 284 sherds obtained from the sites of Yeşilova Höyük and Yassıtepe Höyük, two of the oldest settlements in Western Anatolia, and corresponding to a wide period starting from the Neolithic period to the Chalcolithic and Early Bronze Age (EBA), was studied by attenuated total reflectance (ATR) Fourier transform infrared (FTIR) spectroscopy. FTIR data was used to classify the sherds and state their firing conditions as a result of determining their mineralogical composition using principal component analysis (PCA), multivariate curve resolution (MCR), and hierarchical cluster analysis (HCA). According to the PCA results, the second-order derivative FTIR spectrum with 19 smoothing points in the range of 1300–400 cm−1 was determined to be the most successful model in classifying sherds. As a result of MCR, which allows comparison of standard minerals obtained within the scope of the study, it was determined that kaolinite, illite, chlorite, and some feldspar types were dominant in the sherds. According to HCA analysis, all sherds except five of them exhibited a similar mineralogical structure. According to the derivative spectra, it was seen that the sherds had a composition consisting of kaolinite, illite, hematite, chlorite, and some feldspar types at different rates. Upon examining the data, it becomes evident that the sherds are composed of nearly identical minerals and might have been fired in an oxidizing atmosphere. Even while 279 sherds had many mineralogical traits, N-061, EB-030, N-043, C-068, and N-059 were found to have distinct proportions of comparable minerals, including more kaolinite, illite, calcite, and chlorite. Regarding color, additives, and the amount of ceramic pastes present, these five sherds differ from the others. This suggests that there might have been attempts at manufacture or pots brought in from outside the Yeşilova Höyük-Yassıtepe Höyük sites. The results provided from the derivative spectra showed that all sherds except five of them have been fired at about at temperatures slightly above 450–500 °C due to their relatively low kaolinite character, but below 800 °C because diopside and similar high-temperature minerals could not be detected. The five sherds may have been fired at a temperature of around 450–500 °C due to the presence of a high character of kaolinite, illite, and chlorite. It was understood that the residents of the sites of Yeşilova Höyük and Yassıtepe Höyük produced their ceramics using the same raw materials and the same production methods from the Neolithic period to EBA thanks to FTIR and chemometrics, which are very good tools for analyzing large numbers of sherds with low cost and fast analysis time.

Color Genesis and Chromatography of Yellow Silicified Corals

Color plays a vital role in revealing the formation environment and metasomatic processes of silicified coral. This study investigated the color mechanism and colorimetric characteristics of yellow silicified coral from the aspects of gemology and colorimetry. A Mako G-507C industrial camera, Raman spectroscopy, UV-Vis, EDXRF, and XRD were used for the 16 samples in this study. The results showed that the yellow color of the silicified coral was produced by Fe3+ and influenced by its degree of crystallization. The Raman peaks of all silicified corals were consistent with the standard spectral group peaks of α-quartz, where the yellow part was inferred to be goethite. The peaks at 545 and 505 nm, with a secondary peak near 435 nm in the UV-vis first-order derivative spectrum, were consistent with the presence of hematite and goethite, respectively. The band positions of the second-order derivative spectrum were shown to belong to one single-electron leap 6A1 → (4E;4A1) and one electron pair leap (6A1 +6A1) → (4T1 +4T1). The chroma and lightness were mainly affected by Fe3+. By analyzing the correlation between the Fe content and the characteristic peaks, it was found that an increase in the Fe content led to a red shift in the peak position of the main characteristic peaks, as well as an increase in the hight of the corresponding peaks in the UV-visible first-order derivative spectra. In silicified corals, an increasing crystallinity index is correlated with a decreasing phase proportion of moganite, decreasing Fe content in the bulk, and low chroma.

Conformational Analysis of 1,3-Difluorinated Alkanes.

Fluorine substitution can have a profound impact on molecular conformation. Here, we present a detailed conformational analysis of how the 1,3-difluoropropylene motif (-CHF-CH2-CHF-) determines the conformational profiles of 1,3-difluoropropane, anti- and syn-2,4-difluoropentane, and anti- and syn-3,5-difluoroheptane. It is shown that the 1,3-difluoropropylene motif strongly influences alkane chain conformation, with a significant dependence on the polarity of the medium. The conformational effect of 1,3-fluorination is magnified upon chain extension, which contrasts with vicinal difluorination. Experimental evidence was obtained from NMR analysis, where polynomial complexity scaling simulation algorithms were necessary to enable J-coupling extraction from the strong second-order spectra, particularly for the large 16-spin systems of the difluorinated heptanes. These results improve our understanding of the conformational control toolkit for aliphatic chains, yield simple rules for conformation population analysis, and demonstrate quantum mechanical time-domain NMR simulations for liquid state systems with large numbers of strongly coupled spins.

The Sasa–Satsuma equation with high-order discrete spectra in space-time solitonic regions: soliton resolution via the mixed ∂¯ -Riemann–Hilbert problem

In this paper, we investigate the Cauchy problem of the Sasa–Satsuma (SS) equation with initial data belonging to the Schwartz space. The SS equation is one of the integrable higher-order extensions of the nonlinear Schrödinger equation and admits a 3 × 3 Lax representation. With the aid of the ∂¯ -nonlinear steepest descent method of the mixed ∂¯ -Riemann–Hilbert problem, we give the soliton resolution and long-time asymptotics for the Cauchy problem of the SS equation with the existence of second-order discrete spectra in the space-time solitonic regions.

A Heart Sound Signal Classification Method Based on the Mixed Characteristics of Mel Cepstrum Coefficient and Second-Order Spectrum

A Heart Sound Signal Classification Method Based on the Mixed Characteristics of Mel Cepstrum Coefficient and Second-Order Spectrum

Inversion of Ocean Wave Spectrum from the Multi-frequency HF Radar Sea Echoes with Wind and Swell Coexisting

Recently, the multi-frequency HF radar has attracted much attention. However, only a few works focus on the method of obtaining wave information directly. In this work, a wave spectrum inversion method for multi-frequency HF radar in complex sea conditions is proposed. This method considers the swell component in the ocean, and uses multiple HF radars of different frequencies to obtain the echo Doppler spectra from complex sea conditions with wind and swell coexisting. And then the wave spectrum is derived from the high-energy outer second-order spectra and the corresponding first-order peak of the Doppler spectra. In addition, the proposed method is applied to the simulated sea echo data and the result indicates that the method is effective.

Hyphenation of lipophilic ruthenium(II)-diphosphine core with 5-fluorouracil: an effective metallodrug against glioblastoma brain cancer cells.

Glioblastoma multiforme (GBM) is the most common highly aggressive malignant brain tumor, with a very limited chance for survival post-diagnosis and post-treatment. Despite significant advancement in GBM genomics implicated in molecularly targeted chemotherapies, the prognosis remains poor and requires new drug discovery approaches. We used fluoropyrimidine 5-fluorouracil (5-FU), an antimetabolite anticancer drug conjugated or 'caged' within a lipophilic Ru(II)-diphosphine (dppe) core formulated as [RuII(dppe)2(5-FU)]PF6 (Ru-DPPE-5FU), where dppe = 1,2-bis(diphenylphosphino)ethane, and evaluated its in vitro cytotoxicity in depth with aggressive GBM cells (LN229). The hydrophilic nature of 5-FU limits its passage through the blood-brain barrier (BBB), which prevents its effective accumulation and efficacy for GBM tumors. Herein, we attempted to modulate the lipophilicity of 5-FU by inserting it within a well-designed lipophilic {Ru(dppe)2}-core with anticipated higher efficiency towards GBM. The physicochemical properties of [RuII(dppe)2(5-FU)]PF6 (Ru-DPPE-5FU) were studied using various spectroscopic and analytical techniques. The molecular structure was determined using X-ray crystallography, showing a distorted {RuP4NO} octahedral geometry with bidentate (N, O) binding of 5-FU and its aromatization in the Ru(II)-bound form. The 31P-NMR spectra of Ru-DPPE-5FU showed four closely spaced distinct 31P-signals, indicating four unique chemical environments around P, and the strong coupling constants between them make it a second-order spectrum. The RuII/RuIII redox potential in Ru-DPPE-5FU shifted by ∼0.91 V towards the anodic region as compared to its precursor complex cis-[Ru(dppe)2Cl2] (Ru-DPPE-Cl). DFT-based theoretical calculations have been performed to correlate the experimental electronic absorption spectra and redox behaviours of the complexes. The electrostatic potential (ESP) plots indicate the delocalization of the charge density on the O-/F-atom from the 5-FU ligand towards Ru(II) upon its complexation. The antioxidant properties of all the compounds were quantified by a 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay. The hyphenation of the 5-fluorouracil (5-FU) ligand to the lipophilic {Ru(dppe)2}-core endowed lipophilicity to Ru-DPPE-5FU with higher in vitro cytotoxicity (IC50 = 2.37 μM) against the LN229 GBM cells as compared to the hydrophilic 5-FU, suggesting efficient cellular uptake. Further biological assays indicated that the complex is highly potent in inhibiting significant proliferation and spheroid formation and restricting the migratory potentials of the GBM cells. Increased caspase 3/7 activity and the presence of apoptotic bodies at the center of 3-D GBM spheroids as revealed by AO/EB dual staining indicated a deeper penetration of the lipophilic complex. The Ru-DPPE-5FU complex displayed lower cytotoxicity in HaCaT normal cells (IC50 = 7.27 μM) in comparison to LN229 cancer cells with a selectivity index (S.I.) of ≥3. Overall, the synergism and caging of 5-FU within the hydrophobic {Ru(dppe)2}-core improves the pharmacokinetic profile of Ru-DPPE-5FU as a potent anticancer agent for glioblastoma.

High-accuracy classification and origin traceability of peanut kernels based on near-infrared (NIR) spectroscopy using Adaboost - Maximum uncertainty linear discriminant analysis

Peanut kernels, known for their high nutritional value and palatability, are classified as nut food. In this study, peanut kernel samples from six distinct cities in Shandong Province, China, were examined to categorize and trace their origins. Near-infrared (NIR) spectra of samples were captured using a portable NIR-M-R2 spectrometer. After the application of Savitzky-Golay (SG) filtering, the classification was attempted using principal component analysis (PCA) plus linear discrimination analysis (LDA). Additionally, maximum uncertainty linear discriminant analysis (MLDA) was applied for comparison. A specific number of eigenvectors could respectively maximize the classification accuracies, 81.48% for PCA + LDA and 76.54% for MLDA. In order to further improve the classification accuracies, Adaboost-MLDA was proposed to develop a stronger classifier. This method, after 18 iterations, achieved remarkable effects, achieving a high accuracy of 95.06%. In a similar vein, the enhancement with preprocessing techniques multiplicative scatter correction (MSC) + SG and standard normal variate (SNV) + SG raised accuracies to 98.77% and 97.53%, respectively. The results of classifying first-order and second-order derivative spectra using Adaboost-MLDA were also described, achieving accuracies near 100%. The experiment demonstrates that integrating Adaboost with NIR spectroscopy offers a highly accurate method for peanut kernel classification, promising for practical applications in food quality control.

Improved discrimination of COVID-19 based on data enhancement technology and an information balance feature selection (INB) method

The coronavirus disease (COVID-19) ravaged the world in late 2019 and posed a serious threat to human life and property destruction on a global scale. In this paper, the Wasserstein generative adversarial network with gradient penalty (WGAN-GP) method was selected for balancing the data sample, and an information balance feature selection (INB) method was first proposed to realize the accurate discrimination of COVID-19 saliva samples based on the attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy. The results of the experiment showed that the INB method obtained higher classification accuracy than the traditional feature selection methods in both the original spectrum and the second-order derivative spectrum, especially in the second-order derivative spectrum where all the indexes reached about 85 %. In addition, the combination of WGAN_GP data augmentation and the INB method resulted in an accuracy of 88.7 % for the original spectrum and even 90.6 % for the second-order derivative spectrum. According to these findings, classification research using the WGAN_GP data enhancement model may increase classification accuracy. Additionally, the ability to successfully separate COVID-19 indicates that the INB method to identify spectral data features is a workable method, which also offers a fresh viewpoint on feature selection.

Observational constraints on the second-order primordial power spectrum: Exploring a Continuous Spontaneous Localization inspired inflationary model

Inflation, a period of exponential expansion in the early Universe, is considered an important part of the standard ΛCDM cosmological model, and plays a crucial role in explaining a wide range of current observations. The standard inflationary model predicts a primordial spectrum of fluctuations that is nearly scale-independent, fitting remarkably well the latest observational data. Nevertheless, there is an ongoing discussion surrounding the transition from an initial homogeneous and isotropic quantum state, characterizing the matter fields during inflation, to a classical inhomogeneous/anisotropic one, which gives rise to large-scale structure in the Universe. To tackle this issue, in the present work we explore an inflationary scenario where quantum “collapse” (or reduction) occurs naturally during the evolution of the system; this model is inspired in the so called Continuous Spontaneous Localization (CSL) model. Our present work builds upon previous results by considering the primordial power spectrum up to the second order in the Hubble Flow Functions, where we perform an estimation of the model free parameters. By validating the predictions of the model against observational data, we investigate whether this second-order calculation can explain the slight departure from the power law observed in the scalar spectral running index. We hope this research contributes to the understanding of the quantum-to-classical transition and its implications for cosmology.

Correcting the Contamination of Second-order Spectra: Improving Hα Measurements in Reverberation Mapping Campaigns

Long-term spectroscopic monitoring campaigns on active galactic nuclei (AGNs) provide a wealth of information about its interior structure and kinematics. However, a number of the observations suffer from the contamination of second-order spectra (SOS) which will introduce some undesirable uncertainties at the red side of the spectra. In this paper, we test the effect of SOS and propose a method to correct it in the time domain spectroscopic data using the simultaneously observed comparison stars. Based on the reverberation mapping (RM) data of NGC 5548 in 2019, one of the most intensively monitored AGNs by the Lijiang 2.4 m telescope, we find that the scientific object, comparison star, and spectrophotometric standard star can jointly introduce up to ∼30% SOS for Grism 14. This irregular but smooth SOS significantly affects the flux density and profile of the emission line, while having little effect on the light curve. After applying our method to each spectrum, we find that the SOS can be corrected effectively. The deviation between corrected and intrinsic spectra is ∼2%, and the impact of SOS on time lag is very minor. This method makes it possible to obtain the Hα RM measurements from archival data provided that the spectral shape of the AGN under investigation does not have a large change.

Probing the Inelastic Electron Tunneling via the Photocurrent in a Vertical Graphene van der Waals Heterostructure.

Inelastic electron tunneling (IET), accompanied by energy transfer between the tunneling charge carriers and other elementary excitations, is widely used to investigate the collective modes and quasiparticles in solid-state materials. In general, the inelastic contribution to the tunneling current is small compared to the elastic part and is therefore only prominent in the second derivative of the tunneling current with respect to the bias voltage. Here we demonstrate a direct observation of the IET by measuring the photoresponse in a graphene-based vertical tunnel junction device. Characteristic peaks/valleys are observed in the bias-voltage-dependent tunneling photocurrent at low temperatures, which barely shift with the gate voltage applied to graphene and diminish gradually as the temperature increases. By comparing with the second-order differential conductance spectra, we establish that these features are associated with the phonon-assisted IET. A simple model based on the photoexcited hot-carrier tunneling in graphene qualitatively explains the response. Our study points to a promising means of probing the low-energy elementary excitations utilizing the graphene-based van der Waals (vdW) heterostructures.

Nonlinear frequency response analysis of oxygen reduction reaction on silver in strong alkaline media

Understanding the kinetics of the oxygen reduction reaction on silver in highly concentrated alkaline solutions (e.g., 11 M NaOH) and elevated temperature is of great interest both in terms of technical applications of this reaction (e.g., in chloralkali electrolysis) and in fundamental terms. To gain a better understanding of this process, a nonlinear frequency response analysis was performed in combination with a rotating ring disk analysis. Firstly, the ORR was analyzed under steady-state conditions. The number of exchanged electrons was determined based on rotating ring disk experiments. Secondly, two different dynamic ORR kinetic models have been formulated. These models were used to derive the theoretical first- and second-order frequency response functions (FRFs). Furthermore, the first and second-order FRFs were determined experimentally for both 0.1 M NaOH and 11 M NaOH at different temperatures. In the next step, the kinetic model parameters were numerically estimated from the experimental frequency-domain data. Despite its simplicity, the so-called 1-step model was able to reproduce experimental first- and second-order FRFs in 0.1 M NaOH under different conditions (steady-state potentials and rotation rates). At the same time, this model was able to reproduce the first-order FRF in 11 M NaOH at 60 °C, but not the features observed in the second-order FRF spectra. The more complex 4-step model has shown a better capability in explaining experimental findings, like the first and second-order FRFs, but also experimentally observed change of the Tafel slope as well as the influence of mass transport on the number of exchanged electrons. However, it was also not able to reproduce all experimental features. Possible reasons for these discrepancies are discussed.

Comparative analysis of infrared and electrochemical fingerprints of different medicinal parts of Eucommia ulmoides Oliver

The method of establishing fingerprint profiles of chemical characteristics of herbal medicines is currently one of the most effective methods recognized internationally for controlling the quality of herbal or natural medicines. This work systematically examines geographical differences using the bark and leaves of Eucommia ulmoide as the object of study. Electrochemical fingerprinting and Fourier transform infrared (FTIR) fingerprinting were used to obtain information on different samples. We established a common pattern for the profiles. The results showed that the differences in response positions and intensities of the aspiration fingerprint profiles of the samples from various origins were more obvious and could effectively identify the origins. We combined this with principal component analysis, and the results showed that the differences in chemical composition and content of the samples were correlated with the geographical locations in which they were located. The comparative analysis of the second-order derivative spectra further illustrated the differences in chemical composition and content between samples of different origins.

Seismic wavefield information extraction method based on adaptive local singular value decomposition

The key to seismic data processing is to extract the wavefield information related to the subsurface geological targets. In this paper, we first introduce the low-rank signal priority and the strong energy signal priority characteristics of the Singular Value Decomposition (SVD) method. By analyzing the characteristics of the second-order difference spectrum of typical seismic data, we find two indicative parameters to adaptively select the singular value threshold for extracting reflection and diffraction events. Our proposed Adaptive Local Singular Value Decomposition (ALSVD) method takes advantage of the local analysis, and can effectively extract the weak signals in the seismic data. It is a robust method with good resolution and separation integrity, which is beneficial to seismic migration, seismic imaging, and evaluation of Oil and gas traps. Synthetic examples and field examples demonstrate the effectiveness of this method.

Eco-friendly spectrophotometric methods for determination of remdesivir and favipiravir; the recently approved antivirals for COVID-19 treatment

Remdesivir (REM) and Favipiravir (FAV) are recently approved antivirals prescribed in severely ill COVID-19 patients. Therefore, development of new, simple, rapid, sensitive, and selective methods for analysis of such drugs in their pharmaceutical formulations will be highly advantageous. Herein, we have developed different spectrophotometric methods for analysis of the studied analytes. Method I is based on direct spectrophotometric analysis of REM and FAV in ethanol at λmax 244 and 323 nm, respectively. For simultaneous quantitation of REM and FAV, methods II-V were followed. Method II is based on derivative spectrophotometry in which REM was determined in second-order derivative spectra at 248 nm (the zero-crossing wavelength for FAV), while FAV was measured in first-order derivative spectra at 337 nm (the zero-crossing point for REM). Method III is the dual-wavelength method in which spectral intensities were subtracted at 244–207 nm for REM and at 330–400 nm for FAV. Method IV is the ratio subtraction in which ratio spectra were obtained by a suitable divisor followed by subtraction of intensities at 272–340 nm and 335–222 nm for REM and FAV, respectively. Method V is the derivative ratio method in which the obtained ratio spectra in method IV were converted to first-order derivative and then REM and FAV were recorded at 280 and 340 nm, respectively. Calibration graphs were linear in the ranges of 1–10 µg/mL for REM through all methods and 1–20 µg/mL for FAV in methods I and II, and 2–20 µg/mL by the other methods. The evolved methods were applied to pharmaceutical dosage forms of REM and FAV. All the proposed methods were further applied to human plasma samples containing both drugs with acceptable mean recoveries.

Maturity Stage Discrimination of Camellia oleifera Fruit Using Visible and Near-Infrared Hyperspectral Imaging.

The maturity of Camellia oleifera fruit is one of the most important indicators to optimize the harvest day, which, in turn, results in a high yield and good quality of the produced Camellia oil. A hyperspectral imaging (HSI) system in the range of visible and near-infrared (400–1000 nm) was employed to assess the maturity stages of Camellia oleifera fruit. Hyperspectral images of 1000 samples, which were collected at five different maturity stages, were acquired. The spectrum of each sample was extracted from the identified region of interest (ROI) in each hyperspectral image. Spectral principal component analysis (PCA) revealed that the first three PCs showed potential for discriminating samples at different maturity stages. Two classification models, including partial least-squares discriminant analysis (PLS-DA) and principal component analysis discriminant analysis (PCA-DA), based on the raw or pre-processed full spectra, were developed, and performances were compared. Using a PLS-DA model, based on second-order (2nd) derivative pre-processed spectra, achieved the highest results of correct classification rates (CCRs) of 99.2%, 98.4%, and 97.6% in the calibration, cross-validation, and prediction sets, respectively. Key wavelengths selected by PC loadings, two-dimensional correlation spectroscopy (2D-COS), and the uninformative variable elimination and successive projections algorithm (UVE+SPA) were applied as inputs of the PLS-DA model, while UVE-SPA-PLS-DA built the optimal model with the highest CCR of 81.2% in terms of the prediction set. In a confusion matrix of the optimal simplified model, satisfactory sensitivity, specificity, and precision were acquired. Misclassification was likely to occur between samples at maturity stages two, three, and four. Overall, an HSI with effective selected variables, coupled with PLS-DA, could provide an accurate method and a reference simple system by which to rapidly discriminate the maturity stages of Camellia oleifera fruit samples.