Spectral Classification Research Articles

FNet II: Spectral classification of quasars, galaxies, stars, and broad absorption line (BAL) quasars

FNet II: Spectral classification of quasars, galaxies, stars, and broad absorption line (BAL) quasars

Deep Feature Segmentation Model Driven by Hybrid Convolution Network for Hyper Spectral Image Classification

Deep Feature Segmentation Model Driven by Hybrid Convolution Network for Hyper Spectral Image Classification

AI4LUC: deep learning and automated mask labelling to support land use and land cover mapping in the Cerrado biome

ABSTRACT Land use and land cover (LULC) mapping in the Brazilian Cerrado is challenging due to its vastness, seasonal vegetation changes, and spectral class confusion. Artificial intelligence, particularly deep learning (DL), offers promising solutions through semantic segmentation of Cerrado images, though it traditionally requires manual image annotation. This paper introduces the AI4LUC (Artificial Intelligence for Land Use and Land Cover Classification) method, leveraging DL and automated mask labeling to address LULC in the Brazilian Cerrado. AI4LUC includes pre-processing modules and the Smart Mask©Labelling (SML) module, which combines morphological operations with the CerraNetv3 deep convolutional neural network to automatically generate training masks for DL models like U-Net and DeepLabv3+. Additionally, we present the Cerrado Dataset (CerraData-v3), a high-resolution image dataset with 80,000 labeled samples. Our experiments show the SML module achieved an F1-score of 0.6647 and mIoU of 0.5068, demonstrating significant potential for automated semantic segmentation in this large biome. Furthermore, DeepLabv3+ outperformed U-Net by 69.86%. Source code and dataset: https://github.com/ai4luc/AI4LUC.

Classification of Blazars by Activity Types

We have carried out a spectral classification by the Activity Types for all sample of Blazars from the BZCAT v.5 Catalogue, namely the BZB, BZG, BZQ and BZU type objects. The classification is based on the Sloan Digital Sky Survey (SDSS) homogeneous medium-resolution optical spectroscopy and along with the standard BPT-type diagnostic diagrams, we have applied our newly introduced fine classification scheme with subtypes of AGN and considering many more features. Out of 3561 BZCAT objects, 1363 (38.3 %) having SDSS spectra were classified. After the new classification, 749 (54.9 %) of 1363 objects have changed their optical class.

X-ray AGM activity classes based on SDSS spectra

Sources of X-ray radiation are very interesting and at the same time little studied. Their class mainly includes objects that exhibit a certain activity, such as AGNs, cataclysmic variable stars, white dwarfs, pulsars, etc. Identification of X-ray sources is a rather difficult task, and it is not always possible to detect an optical object. In this work, we tried to carry out a spectral classification of X-ray AGNs and identify activity classes, taking as a basis the ROSAT BSC/FSC catalog of X-ray AGNs, compiled on the basis of low-dispersion spectral plates of the first Byurakan survey. We identified these objects with the SDSS spectral catalog and tried, by carefully studying the spectra, to find out their activity classes, as well as to understand which activity class is more common in these sources.

Activity of the Spectral Class K2V Dwarf HD 189733 in a Hot-Jupiter Type Exoplanet

Activity of the Spectral Class K2V Dwarf HD 189733 in a Hot-Jupiter Type Exoplanet

Soil Organic Carbon Prediction Based on Vis-NIR Spectral Classification Data Using GWPCA-FCM Algorithm.

Soil visible and near-infrared reflectance spectroscopy is an effective tool for the rapid estimation of soil organic carbon (SOC). The development of spectroscopic technology has increased the application of spectral libraries for SOC research. However, the direct application of spectral libraries for SOC prediction remains challenging due to the high variability in soil types and soil-forming factors. This study aims to address this challenge by improving SOC prediction accuracy through spectral classification. We utilized the European Land Use and Cover Area frame Survey (LUCAS) large-scale spectral library and employed a geographically weighted principal component analysis (GWPCA) combined with a fuzzy c-means (FCM) clustering algorithm to classify the spectra. Subsequently, we used partial least squares regression (PLSR) and the Cubist model for SOC prediction. Additionally, we classified the soil data by land cover types and compared the classification prediction results with those obtained from spectral classification. The results showed that (1) the GWPCA-FCM-Cubist model yielded the best predictions, with an average accuracy of R2 = 0.83 and RPIQ = 2.95, representing improvements of 10.33% and 18.00% in R2 and RPIQ, respectively, compared to unclassified full sample modeling. (2) The accuracy of spectral classification modeling based on GWPCA-FCM was significantly superior to that of land cover type classification modeling. Specifically, there was a 7.64% and 14.22% improvement in R2 and RPIQ, respectively, under PLSR, and a 13.36% and 29.10% improvement in R2 and RPIQ, respectively, under Cubist. (3) Overall, the prediction accuracy of Cubist models was better than that of PLSR models. These findings indicate that the application of GWPCA and FCM clustering in conjunction with the Cubist modeling technique can significantly enhance the prediction accuracy of SOC from large-scale spectral libraries.

SERS microscopy as a tool for comprehensive biochemical characterization in complex samples.

Surface enhanced Raman scattering (SERS) spectra of biomaterials such as cells or tissues can be used to obtain biochemical information from nanoscopic volumes in these heterogeneous samples. This tutorial review discusses the factors that determine the outcome of a SERS experiment in complex bioorganic samples. They are related to the SERS process itself, the possibility to selectively probe certain regions or constituents of a sample, and the retrieval of the vibrational information in order to identify molecules and their interaction. After introducing basic aspects of SERS experiments in the context of biocompatible environments, spectroscopy in typical microscopic settings is exemplified, including the possibilities to combine SERS with other linear and non-linear microscopic tools, and to exploit approaches that improve lateral and temporal resolution. In particular the great variation of data in a SERS experiment calls for robust data analysis tools. Approaches will be introduced that have been originally developed in the field of bioinformatics for the application to omics data and that show specific potential in the analysis of SERS data. They include the use of simulated data and machine learning tools that can yield chemical information beyond achieving spectral classification.

Predicting Blooming Day of Cut Lily through Wavelength Reflectance Analysis

Domestic export cut lily flowers are expensive in Japan when they are in bud state that has not yet bloomed and when no leaf yellowing has occurred. Predicting the blooming day of domestic cut lily flowers is essential to increase their commodity value. Thermal imaging, spectroscopic technologies, and hyperspectral cameras have recently been used for quality prediction. This study uses a hyperspectral camera, reflectance of wavelength, and a support vector machine (SVM) to evaluate the predictability of blooming days of cut lily flowers. While examining spectra at wavelengths of 750–900 nm associated with pollination, the resultant reflectance was over 75% during six to four days before blooming and 30% on a blooming day, indicating a decline in their reflectance toward blooming. Furthermore, SVM classification models based on kernel function revealed that the quadratic SVM had the highest accuracy at 84.4%, while the coarse Gaussian SVM had the lowest accuracy at 34.4%. The most crucial wavelength for the quadratic SVM was 842.3 nm, which was associated with water. The quadratic SVM’s accuracy, verified using the area under the curve (ACU), was above 0.8, showing suitability for spectral classification based on blooming day prediction. Thus, this study shows that hyperspectral imaging can classify spectra based on the blooming day, indicating its potential to predict the blooming day, vase life, and quality of cut lily flowers.

QCL Infrared Spectroscopy Combined with Machine Learning as a Useful Tool for Classifying Acetaminophen Tablets by Brand.

The development of new methods of identification of active pharmaceutical ingredients (API) is a subject of paramount importance for research centers, the pharmaceutical industry, and law enforcement agencies. Here, a system for identifying and classifying pharmaceutical tablets containing acetaminophen (AAP) by brand has been developed. In total, 15 tablets of 11 brands for a total of 165 samples were analyzed. Mid-infrared vibrational spectroscopy with multivariate analysis was employed. Quantum cascade lasers (QCLs) were used as mid-infrared sources. IR spectra in the spectral range 980-1600 cm-1 were recorded. Five different classification methods were used. First, a spectral search through correlation indices. Second, machine learning algorithms such as principal component analysis (PCA), support vector classification (SVC), decision tree classifier (DTC), and artificial neural network (ANN) were employed to classify tablets by brands. SNV and first derivative were used as preprocessing to improve the spectral information. Precision, recall, specificity, F1-score, and accuracy were used as criteria to evaluate the best SVC, DEE, and ANN classification models obtained. The IR spectra of the tablets show characteristic vibrational signals of AAP and other APIs present. Spectral classification by spectral search and PCA showed limitations in differentiating between brands, particularly for tablets containing AAP as the only API. Machine learning models, specifically SVC, achieved high accuracy in classifying AAP tablets according to their brand, even for brands containing only AAP.

The Identification of Manure Spreading on Bare Soil through the Development of Multispectral Indices from Sentinel-2 Data: The Emilia-Romagna Region (Italy) Case Study.

Satellite remote sensing is currently an established, effective, and constantly used tool and methodology for monitoring agriculture and fertilisation. At the same time, in recent years, the need for the detection of livestock manure and digestate spreading on the soil is emerging, and the development of spectral indices and classification processes based on satellite multispectral data acquisitions is growing. However, the application of such indicators is still underutilised and, given the polluting impact of livestock manure and digestate on soil, groundwater, and air, an in-depth study is needed to improve the monitoring of this practice. Additionally, this paper aims at exposing a new spectral index capable of detecting the land affected by livestock manure and digestate spreading. This indicator was created by studying the spectral response of bare soil and livestock manure and digestate, using Copernicus Sentinel-2 MSI satellite acquisitions and ancillary datasets (e.g., soil moisture, precipitation, regional thematic maps). In particular, time series of multispectral satellite acquisitions and ancillary data were analysed, covering a survey period of 13 months between February 2022 and February 2023. As no previous indications on fertilisation practices are available, the proposed approach consists of investigating a broad-spectrum area, without investigations of specific test sites. A large area of approximately 236,344 hectares covering three provinces of the Emilia-Romagna Region (Italy) was therefore examined. A series of ground truth points were also collected for assessing accuracy by filling in the confusion matrix. Based on the definition of the spectral index, a value of the latter greater than three provides the most conservative threshold for detecting livestock manure and digestate spreading with an accuracy of 62.53%. Such results are robust to variations in the spectral response of the soil. On the basis of these very encouraging results, it is considered plausible that the proposed index could improve the techniques for detecting the spreading of livestock manure and digestate on bare ground, classifying the areas themselves with a notable saving of energy compared to the current investigation methodologies directly on the ground.

Technical note: A GPU-based shared Monte Carlo method for fast photon transport in multi-energy x-ray exposures.

The Monte Carlo (MC) method is an accurate technique for particle transport calculation due to the precise modeling of physical interactions. Nevertheless, the MC method still suffers from the problem of expensive computational cost, even with graphics processing unit (GPU) acceleration. Our previous works have investigated the acceleration strategies of photon transport simulation for single-energy CT. But for multi-energy CT, conventional individual simulation leads to unnecessary redundant calculation, consuming more time. This work proposes a novel GPU-based shared MC scheme (gSMC) to reduce unnecessary repeated simulations of similar photons between different spectra, thereby enhancing the efficiency of scatter estimation in multi-energy x-ray exposures. The shared MC method selects shared photons between different spectra using two strategies. Specifically, we introduce spectral region classification strategy to select photons with the same initial energy from different spectra, thus generating energy-shared photon groups. Subsequently, the multi-directional sampling strategy is utilized to select energy-and-direction-shared photons, which have the same initial direction, from energy-shared photon groups. Energy-and-direction-shared photons perform shared simulations, while others are simulated individually. Finally, all results are integrated to obtain scatter distribution estimations for different spectral cases. The efficiency and accuracy of the proposed gSMC are evaluated on the digital phantom and clinical case. The experimental results demonstrate that gSMC can speed up the simulation in the digital case by ∼37.8% and the one in the clinical case by ∼20.6%, while keeping the differences in total scatter results within 0.09%, compared to the conventional MC package, which performs an individual simulation. The proposed GPU-based shared MC simulation method can achieve fast photon transport calculation for multi-energy x-ray exposures.

Identification of dominant tree species based on Resource-1 02D hyperspectral image data

Fine-grained classification of tree species by using high-spectral image data has garnered considerable attention from scholars. In this study, through field measurements from Maguan County, Wenshan Prefecture, Yunnan Province, China, high-spectral image data from the Chinese Resource-1 02D satellite were used as the data source. Various analyses were conducted on the original image’s spectral curve, the spectral curve after envelope removal, the spectral curve after first-order differential transformation, and the spectral curve after second-order differential transformation. A spectral angle mapping classification method was employed to classify and identify four dominant tree species in Maguan County, and the accuracy of the classification results was validated using a confusion matrix. Results indicate that the highest accuracy in tree species classification was achieved when first-order differential transformation and envelope removal were used for the spectral curve; the overall accuracy exceeded 95%, and the kappa value was approximately 0.95. The classification results for the spectral curve after second-order differential transformation were the lowest, with an overall accuracy of 81.69% and a kappa value of 0.76. This research demonstrates that applying first-order differential transformation or envelope removal in combination with spectral angle mapping classification considerably reduces data processing time and improves tree species classification accuracy.

Deep jointly optical spectral band selection and classification learning

Spectral data provide material-specific information across a broad electromagnetic wavelength range by acquiring numerous spectral bands. However, acquiring such a significant volume of data introduces challenges such as data redundancy, long acquisition times, and substantial storage capacity. To address these challenges, band selection is introduced as a strategy that focuses on only using the most significant bands to preserve spectral information for a specific task. State-of-the-art methods focus on searching for the most significant bands from previously acquired data, regardless of the optical system and the classification model. Nevertheless, some deep-learning methods, such as end-to-end frameworks, allow the design of optical systems and the learning of the classification network parameters. In this paper, we model the optical band selection as a trainable layer that is coupled with a classification network, where the parameters are learned in an end-to-end framework. To guarantee a physically implementable system, we proposed two regularization terms in the training step to promote binarization and also the number of the selected bands, as we need to provide the conditions to design the physical element where the light passes through. The proposed method provides better performance than state-of-the-art band selection methods for three different spectral datasets under the same conditions.

Using a new spectral disentangling approach to ascertain whether the massive binary HDE 228766 contains a Wolf–Rayet star

ABSTRACT The massive binary HDE 228766 is composed of an O type primary and an evolved secondary. However, previous qualitative analyses of the composite spectrum have led to a wide discussion about whether the secondary is an Of or a Wolf–Rayet star. We use new observations and our novel QER20 package to disentangle for the first time the spectra of the two stellar components and obtain artefact$-$free reconstructed spectra, yielding the more accurate and reliable spectral classifications of O7.5 V((f))z for the primary and O6 Iaf for the secondary. The emission features of the P$-$Cygni profiles of the H $\beta$ and He i 5876 Å lines, present in the reconstructed spectrum of the secondary, show that this star is at an initial phase of its transition to the WN evolutionary stage. A previously unobserved variable emission, composed of at least four independent features, is seen since 2014 superposed to the H $\gamma$ absorption line. Our analysis reveals that these emission features originate from a physically extended region. This could be explained by an episode of enhanced mass-loss in the scenario of a non$-$conservative evolution of the binary.

Power spectral density-based resting-state EEG classification of first-episode psychosis

Historically, the analysis of stimulus-dependent time–frequency patterns has been the cornerstone of most electroencephalography (EEG) studies. The abnormal oscillations in high-frequency waves associated with psychotic disorders during sensory and cognitive tasks have been studied many times. However, any significant dissimilarity in the resting-state low-frequency bands is yet to be established. Spectral analysis of the alpha and delta band waves shows the effectiveness of stimulus-independent EEG in identifying the abnormal activity patterns of pathological brains. A generalized model incorporating multiple frequency bands should be more efficient in associating potential EEG biomarkers with first-episode psychosis (FEP), leading to an accurate diagnosis. We explore multiple machine-learning methods, including random-forest, support vector machine, and Gaussian process classifier (GPC), to demonstrate the practicality of resting-state power spectral density (PSD) to distinguish patients of FEP from healthy controls. A comprehensive discussion of our preprocessing methods for PSD analysis and a detailed comparison of different models are included in this paper. The GPC model outperforms the other models with a specificity of 95.78% to show that PSD can be used as an effective feature extraction technique for analyzing and classifying resting-state EEG signals of psychiatric disorders.

Evaluating the Accuracy of Change detection Using Spectral Indices and spectral classification, A Case Study of Fayoum Governorate, Egypt

Evaluating the Accuracy of Change detection Using Spectral Indices and spectral classification, A Case Study of Fayoum Governorate, Egypt

Identification and Removal of Pollen Spectral Interference in the Classification of Hazardous Substances Based on Excitation Emission Matrix Fluorescence Spectroscopy.

Sensitively detecting hazardous and suspected bioaerosols is crucial for safeguarding public health. The potential impact of pollen on identifying bacterial species through fluorescence spectra should not be overlooked. Before the analysis, the spectrum underwent preprocessing steps, including normalization, multivariate scattering correction, and Savitzky-Golay smoothing. Additionally, the spectrum was transformed using difference, standard normal variable, and fast Fourier transform techniques. A random forest algorithm was employed for the classification and identification of 31 different types of samples. The fast Fourier transform improved the classification accuracy of the sample excitation-emission matrix fluorescence spectrum data by 9.2%, resulting in an accuracy of 89.24%. The harmful substances, including Staphylococcus aureus, ricin, beta-bungarotoxin, and Staphylococcal enterotoxin B, were clearly distinguished. The spectral data transformation and classification algorithm effectively eliminated the interference of pollen on other components. Furthermore, a classification and recognition model based on spectral feature transformation was established, demonstrating excellent application potential in detecting hazardous substances and protecting public health. This study provided a solid foundation for the application of rapid detection methods for harmful bioaerosols.

Superpixel‐guided locality preserving projection and spatial–spectral classification for hyperspectral image

Superpixel‐guided locality preserving projection and spatial–spectral classification for hyperspectral image

The Role of Active Galactic Nucleus Winds in Galaxy Formation: Connecting AGN Outflows at Low Redshifts to the Formation/Evolution of Their Host Galaxies

Using Sloan Digital Sky Survey (SDSS) spectra, we applied an automatic method to search for outflows (OFs) in three large samples of narrow-line active galactic nuclei (AGN) at low redshifts (z < 0.4), separated into three spectral activity classes: radio-loud galaxies (RGs), 15,793; radio-quiet Seyfert 2 AGN (Sy2), 18,585; and LINERs, 25,656. In general, the probability of detecting an OF decreases along the sequence Sy1→Sy2→LINER/RG and independently of the AGN class, the wind velocity, traced by W80, increases with the AGN luminosity. Moreover W80 is systematically higher in RGs or any of the other AGN classes when detected in radio. These results support the idea that there are two main modes of production of OF, the radiative mode dominant in radio-quiet AGN and the jet mode dominant in RGs, although both modes could also happen simultaneously at different levels. From the spectra and SDSS photometry, the characteristics of the AGN host galaxies and their supermassive black holes (SMBHs) were also retrieved using the stellar population synthesis code STARLIGHT. This revealed that, independently of the AGN spectral class, (1) galaxy hosts with OFs have systematically later morphological types and higher star formation rates (SFRs) than their counterparts without OF, (2) the AGN occupy different positions in the specific diagnostic diagram (specific black hole accretion rate (sBHAR) versus specific SFR), which suggests they follow different evolutionary paths congruent with the morphology of their galaxy hosts, and (3) they show no evidence of AGN quenching or triggering of star formation. These results are consistent with a scenario explaining the different AGN classes as consequences of different formation processes of galaxies: early-type galaxies (LINERs and RGs) formed bigger bulges and more massive SMBHs, exhausting their reservoir of gas more rapidly than late-type galaxies (Sy2 and Sy1), and thereby quenching their star formation and starving their SMBHs.