Infrared Data Research Articles

CatBoost algorithm for estimating maize above-ground biomass using unmanned aerial vehicle-based multi-source sensor data and SPAD values

The rapid and accurate estimation of maize above-ground biomass (AGB) is pivotal for precise agricultural management. The rapid evolution of unmanned aerial vehicles (UAVs) and sensor technology has introduced a novel method for obtaining AGB information. Nevertheless, individual sensors may lack comprehensive data, leading to reduced AGB estimation accuracy in certain scenarios. This study collected UAV multi-spectral (MS) and thermal infrared (TIR) data, alongside soil and plant analyzer development (SPAD) values, from maize across multiple growth stages (jointing, trumpet, and big trumpet) during 2022 and 2023. Diverse data fusion programs were devised to explore the potential of combining multi-source sensor data with SPAD values to estimate AGB. The efficacy of CatBoost was evaluated and benchmarked against Support Vector Regression (SVR) and Random Forest Regression (RFR) algorithms. For the entire growth, findings reveal that the fusion of multi-source sensor data (MS + TIR) can mitigate the data insufficiency in single-sensor estimations. The resulting R2 values range from 0.608 to 0.817. Optimal estimation outcomes were achieved by the fusion of multi-source sensor data with SPAD values (MS + TIR + SPAD), yielding R2 values ranging from 0.685 to 0.872. For a single growth stage, there are variations in the estimation accuracy across different growth stages. From the jointing stage to the big trumpet stage, the estimation accuracy consistently increases, with the highest accuracy observed during the big trumpet stage, with R2 ranging from 0.721 to 0.901. Additionally, in alignment with the results for the entire growth stage, the fusion of multi-source sensor data with SPAD values still yields the highest estimation accuracy during different growth stages. In a comparison of different machine learning algorithms, for both the entire growth stage and single growth stages, SVR, RFR, and CatBoost achieved R2 values ranging from 0.305 to 0.824, 0.368 to 0.881, and 0.451 to 0.901, respectively. Notably, the CatBoost algorithm exhibited heightened estimation accuracy. The fusion of multi-source sensor data with SPAD values combined with the CatBoost algorithm results in accurate and reliable maize AGB estimation accuracy. This high-throughput approach to crop phenotyping is characterized by speed and accuracy and serves as a valuable reference for rapidly acquiring AGB information in this geographical region.

A chemistry-based explainable machine learning model based on NIR spectra for predicting wood properties and understanding wavelength selection

ABSTRACT A chemistry-based explainable machine learning (ML) approach was used to predict wood properties using near infrared (NIR) spectral data collected from rough and smooth surfaces, and to provide better understanding of the role of important NIR wavelengths (features) in the performance of ML models. NIR spectra collected from western hemlock (Tsuga heterophylla) and coastal Douglas-fir (Pseudotsuga menziesii) boards with rough and smooth surfaces were fed into random forest and TreeNet; a gradient boosting machine algorithm, for predicting wood density, modulus of elasticity (MOE) and modulus of rupture (MOR). The TreeNet model could predict the MOE, MOR, and density with R2 of 0.66, 0.64, and 0.64 using spectra collected from rough surface and R2 of 0.54, 0.46, and 0.46 using spectra collected from smooth surface. TreeNet outperformed the random forest, and for both ML algorithms higher R2 and lower error were obtained using NIR data collected from rough surfaces. This suggested that for Douglass fir and western hemlock, NIR spectra could be collected on a sawn surface prior to surface planing. However, it is difficult to generalize the impact of surface roughness on the performance of predictive model as different factors (e.g. what constitutes a smooth or rough surface, variability of data set in terms of wood properties) impact the success of predictive models. NIR features having the greatest influence on TreeNet models were examined and consistently had wood chemistry specific band assignments. The most important features occurred in the O-H first overtone, and C–H second overtone regions and a narrow zone (approximately 2400–2500 nm) of the C–H stretch C–C stretch combination region. Important features also differed by property and with surface roughness. Explaining ML model performance using the relative importance of the NIR features showed the importance of wood chemistry related information when developing models, however MOE and MOR TreeNet models based on smooth surface NIR spectra showed an increased importance of water related features. Overall, the chemistry-based explainable machine learning model approach allows for identification of important NIR features, and regions, and aids in understanding how they contribute to the performance of NIR-based wood property predictive models.

Improving estimation of maize leaf area index by combining of UAV-based multispectral and thermal infrared data: The potential of new texture index

Crop leaf area index (LAI) is one of the important indicators to evaluate crop growth and guide field management, and can be used to predict crop yield. Spectral and thermal information extracted by multispectral (MS) and thermal infrared (TIR) sensors mounted on an unmanned aerial vehicle (UAV) can be used for LAI estimation. Image texture is sensitive to the changes in crop surface grayscale or color characteristics, and can be combined with spectral and thermal information to estimate the LAI. But single texture metric has limitations in LAI estimation. Therefore, the purpose of this study is to construct new texture indices based on texture metrics extracted from MS and TIR images, and combined spectral and thermal information to enhance the estimation accuracy of maize LAI. Three replicates of maize experiments under different irrigation treatments were conducted in 2020. The MS and TIR sensors were mounted on a UAV to acquire maize canopy images during critical growth stages and acquire field LAI value of samples synchronously. The LAI estimation models were established using MS data, TIR data, as well as their combination. These models were constructed by Back Propagation Neural Network (BPNN), Partial least squares regression (PLSR), and Random Forest Regression (RFR). Finally, the performance of LAI estimation models was evaluated by the coefficient of determination (R2), root mean square error (RMSE) and relative root mean square error (rRMSE). Results shown that: (i) Among the eight kinds of texture metrics extracted from MS and TIR images, the texture metric mean (MEA) has the best performance. Compared with single texture metrics, texture indices constructed by different metrics has stronger correlation with LAI. (ii) Adding texture indices to estimation models significantly improved model accuracy, especially multispectral three-texture index(MS-TTI) has higher LAI estimation potential than thermal infrared three-texture index(TIR-TTI). (iii) Compared with the use of MS or TIR data alone, the estimation model constructed by combining MS data and TIR data have better performance. The best estimation model obtained by the RFR method (R2 = 0.862, RMSE = 0.246 and rRMSE = 10.20 %) further improved the LAI estimation of maize, with R2 increasing by 6.55 % and 14.48 %, respectively. In conclusion, the combination of MS and TIR data can effectively improve the estimation accuracy of maize LAI, and also provide a feasible method for monitoring crop growth.

A systematic literature review on object detection using near infrared and thermal images

Significant advances have been achieved in object detection techniques using visible images for applications that include military operations, autonomous driving, and security surveillance. However, the quality of visible images suffers from various environmental and illumination conditions resulting in poor detection outcomes. To remedy this, a wide range of new methodologies using visual images together with infrared (IR) images of various wavelengths (including those referred to as thermal images) are being developed and presented to the open literature. Despite this progress, many challenges of object detection still prevail, and it is important to understand them. In this paper, we present a systematic literature review documenting recent advances in object detection using predominantly IR data. We discuss our systematic review process for the identification, filtering, screening, and selection of the relevant methodologies to include in the literature review. The selected methodologies are analyzed and organized into three main groups: (1) object detection in IR images, which includes detection of labeled objects, small target detection, and background subtraction, (2) object detection on multispectral data, and (3) data fusion approaches. Reviewed studies consider different types of objects, environmental conditions, and types of images, particularly in the IR domain. Finally, we discuss some of the key limitations of the current literature and opportunities for future research for improving object detection using both visible and IR data as well as LiDAR and radar data, when applicable.

Unravelling the nuclear dust morphology of NGC 1365: a two-phase polar–RAT model for the ultraviolet to infrared spectral energy distribution

ABSTRACT We present a 3D radiative transfer model for the spectral energy distribution (SED) of NGC 1365, which is a ‘changing look’ Seyfert 1.8 type active galactic nucleus (AGN). The SED from the ultraviolet (UV) to the infrared (IR) is constructed using archival data from the Ultra-Violet Imaging Telescope (UVIT) onboard AstroSat, along with IR data from the literature. The skirt radiative transfer code is used to model the SED and derive the geometry and composition of dust in this AGN. Similar to our earlier SED model of NGC 4151, the nuclear region of NGC 1365 is assumed to contain a ring or disc-like structure concentric to the accretion disc, composed of large (0.1–1 $\mu$m) graphite grains in addition to the two-phase dusty torus made up of interstellar-medium-type grains (Ring And Torus or RAT model). We also include, for the first time, an additional component of dusty wind in the form of a bipolar cone. We carry out a detailed analysis and derive the best-fitting parameters from a χ2 test to be Rin, r = 0.03 pc, σ = 26°, and τtotal = 20 for the assumed ring–torus–polar wind geometry. Our results suggest the presence of hot dust at a temperature T ∼ 1216 K at the location of the ring that absorbs and scatters the incident UV radiation and emits in the near-IR. In the mid-IR, the major contributors are the polar cone and warm dust with T ∼ 914 K at Rin, t = 0.1 pc. Not only are our model radii in agreement with IR interferometric observations, but also our study reiterates the role of high-resolution UV observations in constraining the dust grain size distribution in the nuclear regions of AGN.

Application of a two source energy balance model coupled with satellite based soil moisture and thermal infrared data

Satellite retrievals of thermal infrared data from the Earth’s surface when combined with the surface energy balance equation are widely used to monitor land surface evapotranspiration (ET). Unfortunately thermal infrared based models using this approach have a systematic positive bias with ET during soil water stress conditions. The Two-source Energy Balance model using information on Soil Moisture as another key constraint (TSEB-SM) improves model performance and utility. Here, high resolution satellite soil moisture data from SMAP (Soil Moisture Active Passive) were used to provide soil water boundary conditions for the TSEB-SM model. This SMAP driven TSEB-SM approach was applied in the Heihe river basin in China across a wide range of soil moisture conditions using MODIS thermal infrared land surface temperature (LST) data. The ET estimates from TSEB-SM were validated using flux tower measurements over the main land cover types in the study area and compared with the results from the original TSEB model. The comparison showed that the TSEB-SM approach reduced the RMSE error in latent heat flux estimates by 5 % to 20 % at satellite overpass time, and TSEB-SM outperformed the original TSEB model in daily ET estimates (RMSE values of 1.0 mm/d versus 1.5 mm/d, respectively) across all validation sites. The magnitude of the TSEB-SM improvements were larger early and late in the growing season and in arid areas, consistent with soil moisture improving estimates of ET particularly under times of higher water stress. The use of SMAP satellite soil moisture data with the TSEB-SM model increased the reliability of ET estimation and provides an improved capacity for continuos monitoring of drought effects on the global water cycle.

Evidence for Fine-grained Material at Lunar Red Spots: Insights from Thermal Infrared and Radar Data Sets

Lunar red spots are small spectrally red features that have been proposed to be the result of non-mare volcanism. Studies have shown that a number of red spots are silicic, and are spectrally distinct from both highlands and mare compositions. In this work, we use data from LRO Diviner, Mini-RF, and Arecibo to investigate the material properties of 10 red spots. We create albedo maps using Diviner daytime solar reflectance data to use as an input to our improved thermophysical model, and calculate the rock abundance (RA) and H-parameter values that best fit Diviner nighttime thermal infrared radiance measurements. The H-parameter can be considered analogous to the thermal inertia of the regolith, with a high H-parameter corresponding to low thermal inertia. We find that the red spots generally have low RA, and do not have a uniform H-parameter but contain localized regions of high H-parameter. We additionally find that the red spots have a low circular polarization ratio (CPR) in many of the same locations that show a low RA and high H-parameter. Low RA, high H-parameter, and low CPR indicate a relative lack of rocks larger than ∼10 cm, which is consistent with previous findings of a mantling of fine-grained pyroclastic material for at least three red spots. Areas with high H-parameter but that do not show clear signs of pyroclastics in other data sets may be evidence of previously undiscovered pyroclastics, or could be due to the unique physical properties (e.g., porosity, rock strength/breakdown resistance) of the rocks that make up the red spots.

Evaluation of the Radiometric Calibration of ZY1-02E Thermal Infrared Data

Following the launch of the ZY1-02E satellite, the thermal infrared sensor aboard the satellite experienced alterations in the space environment, leading to varying degrees of attenuation in some components. The laboratory calibration accuracy could not satisfy the demands of quantitative production, and a certain degree of deviation was observed in on-orbit calibration. To accurately characterize the on-orbit radiation properties of thermal infrared remote sensing payloads, an absolute radiometric calibration campaign was carried out at the Ulansuhai Nur and Baotou calibration sites in Inner Mongolia in July 2022. This paper outlines the processes of onboard calibration and vicarious calibration for the ZY1-02E satellite, comparing the outcomes of onboard calibration with those of vicarious calibration. The onboard calibration method involved internal calibration, while the vicarious calibration method utilized an on-orbit absolute radiometric calibration technique based on various natural features that were not constrained by satellite–Earth spectrum matching requirements. Calibration coefficients were acquired, and the absolute radiometric calibration results of on-orbit vicarious and onboard calibration were compared, analyzed, and verified using the radiance computed from measured data and the reference sensor data. The accuracy of on-orbit absolute vicarious calibration was determined, and the causes for the decline in the radiation calibration accuracy on the orbiting satellite were examined. The findings revealed that the vicarious calibration results exhibited a lower percentage of radiance deviation compared with the onboard calibration results, meeting the quantitative requirements of remote sensing data. These results were significantly better than those obtained from onboard blackbody calibration, offering a data foundation for devising satellite calibration plans and enhancing calibration algorithms. In the future, the developmental trend of on-orbit radiometric calibration technology will encompass high-precision and slow-attenuation onboard calibration techniques, as well as high-frequency and simplified-step vicarious calibration methods.

Roadmap for Postnatal Brain Maturation: Changes in Gray and White Matter Composition during Development Measured by Fourier Transformed Infrared Microspectroscopy.

Key events in postnatal brain development, such as neuronal migration, synaptogenesis, and myelination, shape the adult brain. These events are reflected in changes in gray and white matter (GM and WM) occurring during this period. Therefore, precise knowledge of GM and WM composition in perinatal brain development is crucial to characterizing brain formation as well as the neurodevelopmental disruption observed in diseases such as autism and schizophrenia. In this study, we combined histochemical and immunohistochemical staining with biochemical and biophysical analyses using Fourier transform infrared (IR) microspectroscopy (μFTIR) to better understand the chemical changes during postnatal developmental myelination. For this purpose, we analyzed the GM and WM in the mouse brain and cerebellum (strain C57BL/6) from postnatal day 0 (P0) to day P28 and established presumed correlations between staining and IR data. IR spectra allowed the (i) quantification of lipid and protein content through the CH2/amide I ratio, (ii) determination of chemical characteristics of lipids, such as the presence of unsaturated bonds in the carbonate chain or carbonyls from ester groups in the polar head, and (iii) determination of the protein secondary structure (α-helix and intramolecular β-sheets). The results indicate that the increase in the CH2/amide I ratio calculated from the μFTIR data correlates well with lipid histochemical staining. IR data indicated a change in the lipid composition in WM since carbonyl and unsaturated olefinic groups do not increase when lipids accumulate during myelination. Our correlation analysis between IR data and immunohistochemical staining of myelin-associated proteins revealed that myelin oligodendrocyte protein correlated well with lipid accumulation, while myelin basic protein appeared before lipid modifications, which indicated that myelin-associated proteins and lipid deposition were not synchronic. These events were related to a decrease in the intramolecular β/α protein ratio. Our results indicate that lipids and proteins in WM substantially change their composition due to primary myelination, and according to results obtained from staining, these modifications are better described by lipid histochemical staining than by immunohistochemistry against myelin-related proteins. In conclusion, μFTIR can be a useful technique to study WM during perinatal development and provide detailed information about alterations in the chemical composition related to neurodevelopmental diseases.

An Improved UAV-Based ATI Method Incorporating Solar Radiation for Farm-Scale Bare Soil Moisture Measurement

The use of UAV-based remote sensing for soil moisture has developed rapidly in recent decades, with advantages such as high spatial resolution, flexible work arrangement, and ease of operation. In bare and low-vegetation-covered soils, the apparent thermal inertia (ATI) method, which adopts thermal infrared data from UAV-based remote sensing, has been widely used for soil moisture estimation at the field scale. However, the ATI method may not perform well under inconsistent weather conditions due to inconsistency of the intensity of the soil surface energy input. In this study, an improvement of the ATI method (ATI-R), considering the variation in soil surface energy input, was developed by the incorporation of solar radiation measurements. The performances of the two methods were compared using field experiment data during multiple heating processes under various weather conditions. It showed that on consistently sunny days, both ATI-R and ATI methods obtained good correlations with the volumetric water contents (VWC) (R2ATI-R = 0.775, RMSEATI-R = 0.023 cm3·cm−3 and R2ATI = 0.778, RMSEATI = 0.018 cm3·cm−3) on cloudy or a combination of sunny and cloudy days as long as there were significant soil-heating processes despite the different energy input intensities; the ATI-R method could perform better than the ATI method (cloudy: R2ATI-R = 0.565, RMSEATI-R = 0.024 cm3·cm−3 and R2ATI = 0.156, RMSEATI = 0.033 cm3·cm−3; combined: R2ATI-R = 0.673, RMSEATI-R = 0.028 cm3·cm−3 and R2ATI = 0.310, RMSEATI = 0.032 cm3·cm−3); and on overcast days, both the ATI-R and ATI methods could not perform satisfactorily (R2ATI-R = 0.027, RMSEATI-R = 0.024 cm3·cm−3 and R2ATI = 0.027, RMSEATI = 0.031 cm3·cm−3). The results indicate that supplemental solar radiation data could effectively expand applications of the ATI method, especially for inconsistent weather conditions.

V445 Puppis: Dustier than a Thousand Novae

V445 Puppis, the only known Galactic helium nova, is a unique test bed to verify supernova (SN) theories in the single-degenerate channel that involve a white dwarf (WD) accreting matter from a helium-rich donor. An estimate of the mass of the helium shell on the WD is crucial to deciding whether or not it will undergo an SN detonation. In this context, this study estimates the dust and ejecta masses in the 2000 November eruption of V445 Pup. Subsequent to its outburst, the star became cocooned in a dust envelope. An analysis of the spectral energy distribution (SED) of the dust using infrared data shows that V445 Pup produced at least 10−3 M ⊙ of dust, which is unprecedented for a classical or recurrent nova. The SED can be explained by a combination of a cold dust component at 105 ± 10 K, mass (1.9 ± 0.8) × 10−3 M ⊙, and a warm dust component at 255 ± 10 K, mass (2.2 ± 1.2) × 10−5 M ⊙. For a conservative choice of the gas-to-dust mass ratio in the range 10–100, the mass of the ejecta is 0.01–0.1 M ⊙. Such a high mass range raises the question: why did V445 Pup not detonate as a Type Ia SN as is predicted in certain double-detonation sub-Chandrasekhar supernovae formalisms? We reexamine the nature of V445 Pup and discuss its role as a potential SN progenitor.

Remote Sensing Monitoring of Rice Diseases and Pests from Different Data Sources: A Review

Rice is an important food crop in China, and diseases and pests are the main factors threatening its safety, ecology, and efficient production. The development of remote sensing technology provides an important means for non-destructive and rapid monitoring of diseases and pests that threaten rice crops. This paper aims to provide insights into current and future trends in remote sensing for rice crop monitoring. First, we expound the mechanism of remote sensing monitoring of rice diseases and pests and introduce the applications of different commonly data sources (hyperspectral data, multispectral data, thermal infrared data, fluorescence, and multi-source data fusion) in remote sensing monitoring of rice diseases and pests. Secondly, we summarize current methods for monitoring rice diseases and pests, including statistical discriminant type, machine learning, and deep learning algorithm. Finally, we provide a general framework to facilitate the monitoring of rice diseases or pests, which provides ideas and technical guidance for remote sensing monitoring of unknown diseases and pests, and we point out the challenges and future development directions of rice disease and pest remote sensing monitoring. This work provides new ideas and references for the subsequent monitoring of rice diseases and pests using remote sensing.

Water‐surface infrared small object detection based on spatial feature weighting and class balancing method

AbstractInfrared imaging is widely used due to its penetration capability to operate under many weather or lighting condition. However, due to the far distance of aerial view, feature blur, and the scarcity of aerial infrared data, the detection of small infrared targets on the water surface remains a challenging problem. In response to the problem of unclear features, we propose the spatial feature weighting method based on 2D Gaussian distribution. This method increases the weight of the target area by adaptively adjusting the feature activation. Secondly, for the problem of rare aerial perspective infrared data, we propose the cross‐spectral data migration method. By introducing the domain difference loss function to optimize the pseudo‐label selection process, the range of target domain distribution is expanded, and the adaptability of the detector is improved. Finally, in response to the problem of underfitting caused by category imbalance in transfer learning, we propose the class balancing method that effectively reduces the false detection. Extensive experiments were conducted on both benchmark datasets and the self‐built dataset to evaluate the effectiveness and robustness of our method. The proposed method was evaluated with different models and various scenarios, and the results demonstrated the effectiveness.

S-type Stars from LAMOST DR10: Classification of Intrinsic and Extrinsic Stars

In this paper, we found 2939 S-type stars from LAMOST Data Release 10 using two machine-learning methods, and 2306 of them were reported for the first time. The main purpose of this work is to study how to divide S-type stars into intrinsic and extrinsic stars with photometric data and LAMOST spectra. Using infrared photometric data, we adopted two methods to distinguish S-type stars, i.e., the XGBoost algorithm and color–color diagrams. We trained the XGBoost model with 15 input features consisting of colors and absolute magnitudes from Two Micron All Sky Survey (2MASS), AllWISE, AKARI, and IRAS, and found that the model trained by input features with 2MASS, AKARI, and IRAS data has the highest accuracy of 95.52%. Furthermore, using this XGBoost model, we found four color–color diagrams with six infrared color criteria to divide S-type stars, which have an accuracy of about 90%. Applying the two methods to the 2939 S-type stars, 381 (XGBoost)/336 (color–color diagrams) intrinsic and 495 (XGBoost)/82 (color–color diagrams) extrinsic stars were classified, respectively. Using these photometrically classified intrinsic and extrinsic stars, we retrained the XGBoost model with their blue and red medium-resolution spectra, and the 2939 stars were divided into 855 intrinsic and 2056 extrinsic stars from spectra with an accuracy of 94.82%. In addition, we also found the four spectral regions of Zr i (6451.6 Å), Ne ii (6539.6 Å), H α (6564.5 Å), and Fe i (6609.1 Å) and C i (6611.4 Å) are the most important features, which can reach an accuracy of 92.1% when using them to classify S-type stars.

Simulation of visible light at night from infrared measurements using deep learning technique

Improving visible imagery of clouds, especially at night, is of great importance for weather forecasting and warning. This study considered full-disc data obtained by the FengYun-4B satellite with the Advanced Geostationary Radiation Imager (AGRI) as an example, and adopted a deep neural network model to simulate the full-disc visible image using observed dynamic infrared data and surface background data to compensate for missing visible image data at night. The generated daytime visible image had the same characteristics as a corresponding GeoColor image, with an R2 value of 0.99. The simulated night-time visible light, especially in the visualization of low cloud, clearly displayed and highlighted the characteristics of low cloud and fog, and could play an important role in application to near real-time weather forecasting at night. The simulation method could also be applied to the FY-3D/MERSI and Himawari-9/AHI satellite systems and to other infrared bands, similar to the FY-4B/AGRI system.

Monitoring the Mauna Loa (Hawaii) eruption of November–December 2022 from space: Results from GOES-R, Sentinel-2 and Landsat-8/9 observations

Mauna Loa, one of the most actives volcanoes on Earth, is a shield volcano, located on the Island of Hawaii (USA). On 27 November 2022, after about 38 years of quiescence, a new eruptive activity took place at the Moku‘āweoweocaldera, continuing in the following days (i.e. until 10 December) from the fissure vents opening on the Northeast Rift Zone.In this work, we investigate the Mauna Loa November − December 2022 eruption from space, integrating the information from different satellite sensors. The analysis of short-wave infrared (SWIR) data, at 10 min temporal resolution, from the Advanced Baseline Imager (ABI), aboard the Geostationary Operational Environmental Satellites − R series (GOES-R), performed through the Normalised Hotspot Indices (NHI), indicates that the Mauna Loa eruption started on 27 November in between 23:10–23:20 LT (28 November at 09:10–09:20 UTC). The same analysis shows the increase of thermal activity and its progressive reduction from the early morning of 28 November, in agreement with the eruption migration from the summit caldera to the Northeast Rift Zone. By analysing the second phase of eruption through SWIR data from the Multispectral Instrument (MSI) and Operational Land Imager (OLI), respectively aboard Sentinel-2 and Landsat 8/9 satellites, we estimated a maximum lava flow length of 17 km. Moreover, we retrieved values of the volcanic radiative power (VRP) up to 65 GW, and a time-averaged discharge rate (TADR) of ∼1000 (±500) m3/s. These results show that SWIR observations, at different spatial and temporal resolution, may give an important contribution to the monitoring, mapping and characterisation of intense lava effusions.

Areca Yellow Leaf Disease Severity Monitoring Using UAV-Based Multispectral and Thermal Infrared Imagery

The areca nut is the primary economic source for some farmers in southeast Asia. However, the emergence of areca yellow leaf disease (YLD) has seriously reduced the annual production of areca nuts. There is an urgent need for an effective method to monitor the severity of areca yellow leaf disease (SAYD). This study selected an areca orchard with a high incidence of areca YLD as the study area. An unmanned aerial vehicle (UAV) was used to acquire multispectral and thermal infrared data from the experimental area. The ReliefF algorithm was selected as the feature selection algorithm and ten selected vegetation indices were used as the feature variables to build six machine-learning classification models. The experimental results showed that the combination of ReliefF and the Random Forest algorithm achieved the highest accuracy in the prediction of SAYD. Compared to manually annotated true values, the R2 value, root mean square error, and mean absolute percentage error reached 0.955, 0.049, and 1.958%, respectively. The Pearson correlation coefficient between SAYD and areca canopy temperature (CT) was 0.753 (p value < 0.001). The experimental region was partitioned, and a nonlinear fit was performed using CT versus SAYD. Cross-validation was performed on different regions, and the results showed that the R2 value between the predicted result of SAYD by the CT and actual value reached 0.723. This study proposes a high-precision SAYD prediction method and demonstrates the correlation between the CT and SAYD. The results and methods can also provide new research insights and technical tools for botanical researchers and areca practitioners, and have the potential to be extended to more plants.

Chemometric Approach for Profiling of Metabolites of Potential Antioxidant Activity in Apiaceae Species Based on LC-PDA-ESI-MS/MS and FT-NIR

Chemometrics is a tool for data mining and unlocking the door for solving big data queries. Apiaceae is a family species which is commonly cultivated worldwide. Although members of this species are widely used as antioxidant, antibacterial, antifungal, and anti-inflammatory agents, their metabolites profiling remains ambiguous. Based on WHO support, chemometrics has been used in evaluating the quality and authenticity of the herbal products. The objective of this study is to profile and characterize phenolic metabolites in nine species from Egyptian cultivars and three different species of German cultivars from the Apiaceae family using multivariate analysis after LC-PDA-ESI-MS/MS and near infrared spectroscopy data are generated. Principal component analysis was successfully applied to distinguish between the nine Egyptian cultivars and the three German cultivars, and hierarchical cluster analysis also confirmed this distinctive clustering. Partial least square regression (PLS-R) models showed a relationship between phytochemicals and antioxidant activities. The metabolites responsible for the clustering pattern and variables important for projection (VIP) were identified, being twelve amongst nine Egyptian cultivar samples and thirteen amongst the Egyptian cultivar and the German cultivar comparison. The identified VIPs were also correlated with the antioxidant activity using PLS-R. In conclusion, the study showed novelty in the application of hyphenated analytical techniques and chemometrics that assist in quality control of herbal medicine.

Combining Near-Infrared (NIR) Analysis and Modelling as a Fast and Reliable Method to Determine the Authenticity of Agarwood (Aquilaria spp.)

The resinous wood produced by the Aquilaria and Gyrinops species—agarwood—is both rare and highly valuable. It is used in products from perfumes to medicines and has an estimated global market value of $32 billion. As a result, the adulteration and illegal purchasing of agarwood is widespread and of specific concern to enforcement agencies globally. Therefore, it is of interest to have a fast, reliable, and user-friendly method to confirm the authenticity of a sample of agarwood. We investigated the use of near infrared (NIR) data to develop a method that rapidly distinguished between authentic and non-authentic agarwood samples, based upon a soft independent model of class analogy (SIMCA), using software specific to the application of infrared data to material authentication. The model showed a clear distinction between the authentic and non-authentic samples. However, the small values involved led to poor automatic validation results.

New framework for land surface temperature and emissivity inversion based on hyperspectral thermal infrared data

ABSTRACT Hyperspectral thermal infrared (HTIR) data offers enormous potential for land surface temperature (LST) and emissivity (LSE) inversion. Methods to fully use a large amount of spectral information, while avoiding data redundancy, is constantly being explored. Therefore, a new framework for LST and LSE inversion using HTIR data is proposed in this study. First, the stepwise iteration method based on information content was used to select channels that were more sensitive to land surface information. Principal component analysis (PCA) and a two-step machine learning method were then used to retrieve the LST and LSE based on the selected channels. The findings demonstrate that the LST and spectra of the LSE can be obtained simultaneously based on the proposed framework. An LST inversion accuracy of 1.5K and an emissivity inversion accuracy of 0.017 can be achieved. However, an increase in the number of channels does not result in superior outcomes. Satisfactory inversion results can be obtained by selecting only 10–35 channels with the highest information content. In conclusion, this framework can be used to guide channel selection and land surface parameters inversion for HTIR sensors.