Portable Spectroradiometer Research Articles

Integration of VIS–NIR Spectroscopy and Multivariate Technique for Soils Discrimination Under Different Land Management

Proximal sensing has become increasingly popular due to developments in soil observation technologies and the demands of timely information gathering through contemporary methods. By utilizing the morphological, physical, and chemical characteristics of representative pedogenetic profiles established in various soils of the Sohag governorate, Egypt, the current research addresses the characterization of surface reflectance spectra and links them with the corresponding soil classification. Three primary areas were identified: recently cultivated, old cultivated, and bare soils. For morphological analysis, a total of 25 soil profiles were chosen and made visible. In the dark room, an ASD Fieldspec portable spectroradiometer (350–2500 nm) was used to measure the spectrum. Based on how similar their surface spectra were, related soils were categorized. Ward’s method served as the basis for the grouping. Despite the fact that the VIS–NIR spectra of the surface soils from various land uses have a similar reflectance shape, it is still possible to compare the soil reflectance curves and the effects of the surface soils. As a result, three groups of soil curves representing various land uses were observed. Cluster analysis was performed on the reflectance data in four ranges (350–750, 751–1150, 1151–1850, and 1851–2500 nm). The groups derived from the soil surface ranges of 350–750 nm and 751–1150 nm were not the same as those derived from the ranges of 1151–1850 nm and 1851–2500 nm. The last two categories are strikingly comparable to various land uses with marginally similar features. Based on the ranges of 1151–1850 nm and 1851–2500 nm in surface spectral data, the dendrogram effectively separated and combined the profiles into two separate clusters. These clusters matched different land uses exactly. The results can be used to promote the widespread usage of in situ hyperspectral data sets for the investigation of various soil characteristics.

Leaf spectrum analysis of three tropical timber species: Diomate (Astronium graveolens), Choibá (Dipteryx oleifera), and Algarrobo (Hymenaea courbaril)

This study analyzed the leaf spectral response of three native timber forest species in the tropical dry forest: Diomate (Astronium graveolens Jacq.), Choibá (Dipteryx oleifera Benth.), and Algarrobo (Hymenaea courbaril L.). The study was conducted at the León Morales Soto Arboretum and Palmetum, at the Universidad Nacional de Colombia in Medellín, Antioquia, Colombia. Spectral data from the leaves were collected in situ using the portable spectroradiometer ASD FieldSpec HandHeld-2, which operates with a spectral resolution of 1 nm (resampled to 10 nm) and covers a spectral range between 325 and 1,075 nm (limited to 400-900 nm). Based on the measurements, the behavior and spectral variability of the species were evaluated. One-factor Analysis of Variance and Mann Whitney-Wilcoxon U-test were implemented in reflectance spectra to select the optimal narrow bands for species discrimination. The classification capacity of the selected narrow bands was assessed using the K-nearest neighbors’ algorithm. It was found that A. graveolens and H. courbaril exhibited spectral signatures typical of healthy vegetation, while D. oleifera showed spectral changes during the early stages of senescence. Regarding spectral separability, 23 narrow bands in the visible region and near-infrared region were identified as optimal for distinguishing the plant species. The supervised classification algorithm applied to these 23 narrow bands achieved an overall accuracy of 95.8%. In conclusion, these findings provide valuable insights into the spectral response of important tropical species and contribute to their conservation efforts by enhancing understanding of their unique spectral characteristics in diverse and heterogeneous ecosystems like tropical forests.

Estimation of Surface Soil Nutrient Content in Mountainous Citrus Orchards Based on Hyperspectral Data

Monitoring soil conditions is of great significance for guiding fruit tree production and increasing yields. Achieving a rapid determination of soil physicochemical properties can more efficiently monitor soil conditions. Traditional sampling and survey methods suffer from slow detection speeds, low accuracy, limited coverage, and require a large amount of manpower and resources. In contrast, the use of hyperspectral technology enables the precise and rapid monitoring of soil physicochemical properties, playing an important role in advancing precision agriculture. Yuxi City, Yunnan Province, was selected as the study area; soil samples were collected and analyzed for soil organic matter (SOM), total nitrogen (TN), total phosphorus (TP), and available nitrogen (AN) contents. Additionally, soil spectral reflectance was obtained using a portable spectroradiometer. Hyperspectral characteristic bands for soil nutrients were selected from different spectral preprocessing methods, and different models were used to predict soil nutrient content, identifying the optimal modeling approach. For SOM prediction, the second-order differentiation-multiple stepwise regression (SD-MLSR) model performed exceptionally well, with an R2 value of 0.87 and RMSE of 6.61 g·kg−1. For TN prediction, the logarithm of the reciprocal first derivative-partial least squares regression (LRD-PLSR) model had an R2 of 0.77 and RMSE of 0.37 g·kg−1. For TP prediction, the logarithmic second-order differentiation-multiple stepwise regression (LTSD-MLSR) model had an R2 of 0.69 and RMSE of 0.04 g·kg−1. For AN prediction, the logarithm of the reciprocal second derivative-partial least squares regression (LRSD-PLSR) model had an R2 of 0.83 and RMSE of 24.12 mg·kg−1. The results demonstrate the high accuracy of these models in predicting soil nutrient content.

Reconstructing fresh green leaf spectra in the SWIR-2 region (2001–2500 nm) collected in a humid environment by referring to publicly available green leaf spectral databases

Leaf spectra (reflectance and transmittance) are commonly measured using a portable spectroradiometer and an integrating sphere or contact probe with an artificial light source. However, spectral data may be obscured due to water vapor and low signal-to-noise ratios, especially in the shortwave infrared-2 region (SWIR-2, 2001–2500 nm). Therefore, we proposed a spectral reconstruction approach to retrieve noise-free SWIR-2 fresh green leaf spectra by referring to the previously published quality-controlled fresh green leaf spectral reference databases. We processed 896 pairs of fresh tea (Camellia sinensis var. sinensis) leaf reflectance/transmittance from Alishan in central Taiwan. We selected a subset (500–1900 nm) of the spectra in the visible, near-infrared, and SWIR-1 regions (VNS-1) that were relatively insensitive to atmospheric conditions. We matched those spectra with publicly available reference green leaf spectral databases, and selected the one that was most similar to each Alishan VNS-1 spectrum. We then used multivariable linear regression, linear parameter multiplication and spectral reversion to reconstruct SWIR-2 spectra. Finally, we used another set of green leaf spectral databases to assess the performance of the proposed method. The performance of the reconstruction approach was satisfactory, with mean (± standard deviation) root-mean-square errors (RMSEs) of 0.0041 ± 0.0019 (reflectance) and 0.0054 ± 0.0027 (transmittance) for each spectrum and RMSEs of 0.0058 ± 0.0027 (reflectance) and 0.0055 ± 0.0043 (transmittance) for each SWIR-2 band. The proposed approach successfully modeled SWIR-2, which could be further improved with the availability of a more comprehensive set of green leaf reference spectral databases.

Prospects for detecting alternariose in potato plantings using hyperspectral measurements

The harmfulness of Alternaria leaf blight on potatoes has increased significantly in recent years in the North-West of Russia. For effective protection of potato plantings from the disease it is extremely important to detect it promptly. For that purpose the ground-based portable spectroradiometer PSR-1100 with a measurement range from 320 to 1100 nm was used. The study was conducted at the experimental station of Menkovo branch of Agrophysical research institute located in Gatchina district of Leningrad region. The result of the study revealed that even in case with a slight damage of potato plants with Alternaria leaf blight (less than 10% of the leaf surface) there were significant changes in the values of the reflectivity of potato leaves were recorded by the device in the near infrared part of the spectrum especially with severe damage to the leaf surface (25-50% of the leaf surface). At the same time there were no significant differences in the entire measured wavelength range (320-1100 nm) in plants with Alternaria leaf blight damage to 25% of the leaf surface. Thus, the detection of Alternaria leaf blight in potato plantings is practicable by means of ground-based hyperspectral measurements.

Investigation of the Physiological Effects of Plant Conditioners in Field Experiments of Winter Wheat

Our study aimed at to test the effects of two different plant conditioners on some morphological parameters, yield and grain quality of winter wheat (Babona) in a field experiment with medium plots, and to investigate some physiological parameters closely related to yield quantity and quality at the beginning of ear emergence by in vivo field measurements. At the beginning of ear emergence we determined the relative chlorophyll content of leaves (SPAD value), moreover leaf reflectance was measured with a portable spectroradiometer to produce spectral vegetation indices that can be used to evaluate the effects of treatments on chlorophyll, carotenoid, anthocyanin and water content of leaves, and to infer photochemical efficiency and stress sensitivity. The results were strongly influenced by the unfavourable rainfall distribution during the growing season. Precipitation deficit in spring significantly reduced the development of the crop stand (yield was below the national and county levels), and its effect was also observable in quality parameters (low raw protein and gluten content), but the positive effect of treatment 1 was detectable: higher yield and quality, higher plant height and ear length compared to the control. At the beginning of ear emergence, some spectral vegetation indices indicated the positive effect of treatment 1 despite the drought: chlorophyll content and photochemical activity of leaves were higher, and higher stress sensitivity and protective pigment concentration in the control.

Non-Destructive Quality Evaluation of 80 Tomato Varieties Using Vis-NIR Spectroscopy.

Traditional biochemical methods are resource- and time-consuming; therefore, there is a need for cost-effective alternatives. A spectral analysis is one of the non-destructive techniques that are more widely used for fruit quality determination; however, references are needed for traditional methods. In this study, visible and near-infrared (Vis-NIR) spectroscopy was used to analyze the internal quality attributes of tomatoes. For the first time, 80 varieties with large differences in fruit size, shape, color, and internal structure were used for an analysis. The aim of this study was to develop models suitable to predict a taste index, as well as the content of lycopene, flavonoids, β-carotene, total phenols, and dry matter of intact tomatoes based on Vis-NIR reflectance spectra. The content of phytochemicals was determined in 80 varieties of tomatoes. A total of 140 Vis-NIR reflectance spectra were obtained using the portable spectroradiometer RS-3500 (Spectral Evolution Inc.). Partial least squares regression (PLS) and multiple scatter correction (MSC) were used to develop calibration models. Our results indicated that PLS models with good prediction accuracies were obtained. The present study showed the high capability of Vis-NIR spectroscopy to determine the content of lycopene and dry matter of intact tomatoes with a determination coefficient of 0.90 for both parameters. A regression fit of R2 = 0.86, R2 = 0.84, R2 = 0.82, and R2= 0.73 was also achieved for the taste index, flavonoids, β-carotene, and total phenols, respectively.

Spectral characterization and severity assessment of rice blast disease using univariate and multivariate models.

Rice is the staple food of more than half of the population of the world and India as well. One of the major constraints in rice production is frequent occurrence of pests and diseases and one of them is rice blast which often causes yield loss varying from 10 to 30%. Conventional approaches for disease assessment are time-consuming, expensive, and not real-time; alternately, sensor-based approach is rapid,non-invasive and can be scaled up in large areas with minimum time and effort. In the present study, hyperspectral remote sensing for the characterization and severity assessment of rice blast disease was exploited. Field experiments were conducted with 20 genotypes of rice having sensitive and resistant cultivars grown under upland and lowland conditions at Almora, Uttarakhand, India. The severity of the rice blast was graded from 0 to 9 in accordance to International Rice Research Institute (IRRI). Spectral observations in field were taken using a hand-held portable spectroradiometer in range of 350-2500 nm followed by spectral discrimination of different disease severity levels using Jeffires-Matusita (J-M) distance. Then, evaluation of 26 existing spectral indices (r≥0.8) was done corresponding to blast severity levels and linear regression prediction models were also developed. Further, the proposed ratio blast index (RBI) and normalized difference blast index (NDBI) were developed using all possible combinations of their correlations with severity level followed by their quantification to identify the best indices. Thereafter, multivariate models like support vector machine regression (SVM), partial least squares (PLS), random forest (RF), and multivariate adaptive regression spline (MARS) were also used to estimate blast severity. Jeffires-Matusita distance was separating almost all severity levels having values >1.92 except levels 4 and 5. The 26 prediction models were effective at predicting blast severity with R2 values from 0.48 to 0.85. The best developed spectral indices for rice blast were RBI (R1148, R1301) and NDBI (R1148, R1301) with R2 of 0.85 and 0.86, respectively. Among multivariate models, SVM was the best model with calibration R2=0.99; validation R2=0.94, RMSE=0.7, and RPD=4.10. The methodology developed paves way for early detection and large-scale monitoring and mapping using satellite remote sensors at farmers' fields for developing better disease management options.

Unmanned Aerial Vehicle (UAV)–Based Imaging Spectroscopy for Predicting Wheat Leaf Nitrogen

Quantitative estimation of crop nitrogen is the key to site-specific management for enhanced nitrogen (N) use efficiency and a sustainable crop production system. As an alternate to the conventional approach through wet chemistry, sensor-based noninvasive, rapid, and near-real-time assessment of crop N at the field scale has been the need for precision agriculture. The present study attempts to predict leaf N of wheat crop through spectroscopy using a field portable spectroradiometer (spectral range of 400–2500 nm) on the ground in the crop field and an imaging spectrometer (spectral range of 400–1000 nm) from an unmanned aerial vehicle (UAV) with the objectives to evaluate (1) four multivariate spectral models (i.e., artificial neural network, extreme learning machine [ELM], least absolute shrinkage and selection operator, and support vector machine regression) and (2) two sets of hyperspectral data collected from two platforms and two different sensors. In the former part of the study, ELM outperforms the other methods with maximum calibration and validation R2 of 0.99 and 0.96, respectively. Furthermore, the image data set acquired from UAV gives higher performance compared to field spectral data. Also, significant bands are identified using stepwise multiple linear regression and used for modeling to generate a wheat leaf N map of the experimental field.

APRENDIZAGEM DE MÁQUINA PARA IDENTIFICAÇÃO DE PLANTAS DE SOJA SOB ATAQUE DE INSETOS USANDO DADOS HIPERESPECTRAIS

The integration between the areas of remote sensing and machine learning has allowed an advance in the way of mapping agricultural fields and monitoring crops. This work investigates the ability of machine learning algorithms to classify soybean plants under insect attack, using reflectance spectroscopy measurements collected at the leaf level. To this end, tests were developed with different algorithms using a set of 991 spectral curves referring to healthy soybean plants under attack by pests, collected in eight consecutive days. These curves were measured by the EMBRAPA team, using a portable spectroradiometer, which records in the range of 350 to 2500 nm. Such curves were, initially, pre-processed to remove the regions of atmospheric absorption by water vapor, and then subdivided into a set of training, validation and testing of the machine learning algorithms. The Google Collabs interpreter was used and the algorithms were written in Python language, using libraries such as Skit Sklearn. Among the algorithms used, there are Random Forest, Decision Tree, Support Vector Machine, Logistic Regression and Extra-Tree. The Extra-tree has better performance (F1-score = 80.40%; precision = 81%; recall = 80%) in the proposed task. It is concluded that it is possible to process reflectance spectroscopy measurements with machine learning algorithms to monitor insect attack on soybean plants. It is recommended that the applied approach be tested in other cultures.

Artificial intelligence applications in precision agriculture to predict the effect of Root-Knot nematodes and grafting on vegetable crop health from proximal remote sensing machines

In precision agriculture, the Normalized Difference Vegetative Index (NDVI) considers the spectral characteristics of healthy green vegetation. This index is an effective way of detecting the green state of plants. This is why we choose to use NDVI as a reference index to predict the effect of Root-Knot nematodes and grafting on vegetable crop health from proximal remote sensing machines. These machines were used to estimate different physiological, biochemical, and agronomic parameters as indicators of stress (GA, GGA, SPAD, and canopy temperature). Leaf level pigments were measured using a handheld sensor (SPAD). Canopy vigor and biomass were assessed using vegetation indices derived from RGB images and the NDVI was measured with a portable spectroradiometer (Greenseeker). The plant level water stress was assessed indirectly by plant temperature using an infrared thermometer. We conclude that the grafted plants were less stressed and more protected against nematode attack. The comparison of NDVI index predicted by AI models showed that artificial neural network MLP demonstrated the best prediction performance than the linear regression method. However, their R-squared decreased from 0.820 to 0.772, and NRMSE increased from 12.3% to 12.4%, respectively.

Using Machine Learning for Nutrient Content Detection of Aquaponics-Grown Plants Based on Spectral Data

Nutrients derived from fish feed are insufficient for optimal plant growth in aquaponics; therefore, they need to be supplemented. Thus, estimating the amount of supplementation needed can be achieved by looking at the nutrient contents of the plant. This study aims to develop trustworthy machine learning models to estimate the nitrogen (N), phosphorus (P), and potassium (K) contents of aquaponically grown lettuce. A FieldSpec4, Pro FR portable spectroradiometer (ASD Inc., Analytical Spectral Devices Boulder, Boulder, CO, USA) was used to measure leaf reflectance spectra, and 128 lettuce seedlings given four NPK treatments were used for spectra acquisition and total NPK estimation. Principal component analysis (PCA), genetic algorithms (GA), and sequential forward selection (SFS) were applied to select the optimal wavebands. Partial least squares regression (PLSR), back-propagation neural network (BPNN), and random forest (RF) approaches were used to develop the predictive models of NPK contents using the selected optimal wavelengths. Good and significantly correlated predictive accuracy was obtained in comparison with the laboratory-measured freshly cut lettuce leaves with R2 ≥ 0.94. The proposed approach provides a pathway toward automatic nutrient estimation of aquaponically grown lettuce. Consequently, aquaponics will become more intelligent, and will be adopted as a precision agriculture technology.

End-of-Life Textile Recognition in a Circular Economy Perspective: A Methodological Approach Based on Near Infrared Spectroscopy

The life cycle of textiles (i.e., fabrics and apparel products) generates many environmental impacts, such as resource consumption, water, soil, and air pollution through the dispersion of chemical substances and greenhouse gases. For these reasons, in 2019, textiles were identified as a “priority product category for the circular economy” by the European Commission that proposed a new circular economy action plan focusing on recycling. An in-depth characterization of textile fabrics could lead to an ad hoc recycling procedure, reducing resource consumption and chemicals utilization. In this work, NIR (1000–1650 nm) spectroscopy was applied to extract information regarding fabric composition, with reference to cotton, silk, viscose, and some of their blends, using two different devices: a hyperspectral imaging (HSI) platform and a portable spectroradiometer. The different fabrics were correctly classified based on their spectral features by both detection instruments. The proposed methodological approach can be applied for quality control in the textile recycling sector at industrial and/or laboratory scale thanks to the easiness of use and the speed of detection.

Comparison of Proximal Remote Sensing Devices of Vegetable Crops to Determine the Role of Grafting in Plant Resistance to Meloidogyne incognita

Proximal remote sensing devices are novel tools that enable the study of plant health status through the measurement of specific characteristics, including the color or spectrum of light reflected or transmitted by the leaves or the canopy. The aim of this study is to compare the RGB and multispectral data collected during five years (2016–2020) of four fruiting vegetables (melon, tomato, eggplant, and peppers) with trial treatments of non-grafted and grafted onto resistant rootstocks cultivated in a Meloidogyne incognita (a root-knot nematode) infested soil in a greenhouse. The proximal remote sensing of plant health status data collected was divided into three levels. Firstly, leaf level pigments were measured using two different handheld sensors (SPAD and Dualex). Secondly, canopy vigor and biomass were assessed using vegetation indices derived from RGB images and the Normalized Difference Vegetation Index (NDVI) measured with a portable spectroradiometer (Greenseeker). Third, we assessed plant level water stress, as a consequence of the root damage by nematodes, using stomatal conductance measured with a porometer and indirectly using plant temperature with an infrared thermometer, and also the stable carbon isotope composition of leaf dry matter.. It was found that the interaction between treatments and crops (ANOVA) was statistically different for only four of seventeen parameters: flavonoid (p < 0.05), NBI (p < 0.05), NDVI (p < 0.05) and the RGB CSI (Crop Senescence Index) (p < 0.05). Concerning the effect of treatments across all crops, differences existed only in two parameters, which were flavonoid (p < 0.05) and CSI (p < 0.001). Grafted plants contained fewer flavonoids (x¯ = 1.37) and showed lower CSI (x¯ = 11.65) than non-grafted plants (x¯ = 1.98 and x¯ = 17.28, respectively, p < 0.05 and p < 0.05) when combining all five years and four crops. We conclude that the grafted plants were less stressed and more protected against nematode attack. Leaf flavonoids content and the CSI index were robust indicators of root-knot nematode impacts across multiple crop types.

Reduction in Blockage Property of UV-Blocking Greenhouse Covering Material: In Situ and Lab Measurement Comparison

The goal of this research was to compare and evaluate the measurements taken by different instruments regarding alterations while varying the ultraviolet (UV)-blocking property of cladding material during its usage under real greenhouse conditions. The UV-blocking covering material, low-density polyethylene (LDPE), is enriched with additives that are scattered in several layers during the manufacturing process, resulting in the reinforcement of its properties mechanically as well as optically. The duration of this study was three years, and the instruments used were: (a) sensors measuring the UV radiation reaching the earth’s surface in its A and B components; and (b) a portable spectroradiometer capable of measuring the transmissivity of a material, only in the UV-A region. Three covering materials were used with different UV radiation transmissivity. The transmittance was measured both in the laboratory (on samples taken from the roof) and in the field (where the greenhouses were located). Equations were defined to describe the variation in UV radiation transmission increase rate as a function of field exposure time. Lastly, it is important to note that the specific UV radiation sensors were extremely accurate.

Estimating Soil Properties and Nutrients by Visible and Infrared Diffuse Reflectance Spectroscopy to Characterize Vineyards

Visible, near, and shortwave infrared (VIS-NIR-SWIR) reflectance spectroscopy, a cost-effective and rapid means of characterizing soils, was used to predict soil sample properties for four vineyards (central and north-western Spain). Sieved and air-dried samples were measured using a portable spectroradiometer (350–2500 nm) and compared for pistol grip (PG) versus contact probe (CP) setups. Raw data processed using standard normal variate (SVN) and detrending transformation (DT) were grouped into four subsets (VIS: 350–700 nm; NIR: 701–1000 nm; SWIR: 1001–2500 nm; and full range: 350–2500 nm) in order to identify the most suitable range for determining soil characteristics. The performance of partial least squares regression (PLSR) models in predicting soil properties from reflectance spectra was evaluated by cross-validation. The four spectral subsets and transformed reflectances for each setup were used as PLSR predictor variables. The best performing PLSR models were obtained for pH, electrical conductivity, and phosphorous (R2 values above 0.92), while models for sand, nitrogen, and potassium showed moderately good performances (R2 values between 0.69 and 0.77). The SWIR subset and SVN + DT processing yielded the best PLSR models for both the PG and CP setups. VIS-NIR-SWIR reflectance spectroscopy shows promise as a technique for characterizing vineyard soils for precision viticulture purposes. Further studies will be carried out to corroborate our findings.

In vivo non-invasive near-infrared spectroscopy distinguishes normal, post-stroke, and botulinum toxin treated human muscles

In post-stroke hemiparesis, neural impairment alters muscle control, causing abnormal movement and posture in the affected limbs. A decrease in voluntary use of the paretic arm and flexed posture during rest also induce secondary tissue transformation in the upper limb muscles. To obtain a specific, accurate, and reproducible marker of the current biological status of muscles, we collected visible (VIS) and short-wave Infrared (SWIR) reflectance spectra in vivo using a portable spectroradiometer (350–2500 nm), which provided the spectral fingerprints of the elbow flexors and extensors. We compared the spectra for the affected and unaffected sides in 23 patients with post-stroke hemiparesis (25–87 years, 8 women) and eight healthy controls (33–87 years, 5 women). In eight patients, spectra were collected before and after botulinum toxin injection. Spectra underwent off-line preprocessing, principal component analysis, and partial least-squares discriminant analysis. Spectral fingerprints discriminated the muscle (biceps vs. triceps), neurological condition (normal vs. affected vs. unaffected), and effect of botulinum toxin treatment (before vs. 30 to 40 days vs. 110 to 120 days after injection). VIS-SWIR spectroscopy proved valuable for non-invasive assessment of optical properties in muscles, enabled more comprehensive evaluation of hemiparetic muscles, and provided optimal monitoring of the effectiveness of medication.

Detection of Brominated Plastics from E-Waste by Short-Wave Infrared Spectroscopy

In this work, the application of Short-Wave Infrared (SWIR: 1000–2500 nm) spectroscopy was evaluated to identify plastic waste containing brominated flame retardants (BFRs) using two different technologies: a portable spectroradiometer, providing spectra of single spots, and a hyperspectral imaging (HSI) platform, acquiring spectral images. X-ray Fluorescence (XRF) analysis was preliminarily performed on plastic scraps to analyze their bromine content. Chemometric methods were then applied to identify brominated plastics and polymer types. Principal Component Analysis (PCA) was carried out to explore collected data and define the best preprocessing strategies, followed by Partial Least Squares—Discriminant Analysis (PLS-DA), used as a classification method. Plastic fragments were classified into “High Br content” (Br > 2000 mg/kg) and “Low Br content” (Br < 2000 mg/kg). The identified polymers were acrylonitrile butadiene styrene (ABS) and polystyrene (PS). Correct recognition of 89–90%, independently from the applied technique, was achieved for brominated plastics, whereas a correct recognition ranging from 81 to 89% for polymer type was reached. The study demonstrated as a systematic utilization of both the approaches at the industrial level and/or at laboratory scale for quality control can be envisaged especially considering their ease of use and the short detection response.

Nitrogen content diagnosis of apple trees canopies using hyperspectral reflectance combined with PLS variable extraction and extreme learning machine

Nitrogen (N) is an important mineral element in apple production. Rapid estimation of apple tree N status is helpful for achieving precise N management. The objective of this work was to explore partial least squares (PLS) regression in dimensional reduction of spectral data and build the diagnostic model. The spectral reflectance data were collected from Fuji apple trees with 4 levels of N fertilizer treatment in the Loess Plateau in 2018 and 2019 using an ASD portable spectroradiometer, and leaf total N content was obtained at the same time. The raw spectra were pretreated using Savitzky-Golay (SG) smoothing and a combination of SG and first-order derivative (SG_FD) or second-order derivative (SG_SD). The samples were divided into a calibration dataset and a prediction dataset using SPXY. Based on 4 factors of PLS regression, including latent variables (LVs), X-loading, variable importance in projection (VIP) and regression coefficients (RC), the 6 methods (LVs, X-loading, VIP_01, VIP_02, RC_01 and RC_02) were derived and used for variable extraction, based on which PLS model and ELM model were established. The results indicated that the spectral data processed by SG_FD had the highest signal-to-noise ratio and was selected for subsequent analysis. The amounts of variables extracted by LVs, X-loading, VIP_01, VIP_02, RC_01 and RC_02 were 6, 11, 18, 305, 26 and 88, respectively. The method of extracting variables with an RC threshold based on the minimum RMSEP (RC_02) could effectively avoid the omission of effective information. The RC_02 method was recommended for related research which required accurate wavelength information as a variable. The variable extraction method based on LVs generated an ELM model with a simple structure. The prediction results showed that the ELM model outperformed the PLS model. The PLS(LVs)_ELM model was the best; R2P, RMSEP and RPD were 0.837, 2.393 and 2.220, respectively. Keywords: partial least square, variable extraction method, extreme learning machine, hyperspectral reflectance, apple tree, canopy nitrogen content DOI: 10.25165/j.ijabe.20211403.6157 Citation: Chen S M, Ma L H, Hu T T, Luo L H, He Q, Zhang S W. Nitrogen content diagnosis of apple trees canopies using hyperspectral reflectance combined with PLS variable extraction and extreme learning machine. Int J Agric & Biol Eng, 2021; 14(3): 181–188.

Hyperspectral sensor: A handy tool to evaluate the efficacy of cleaning procedures

• ASD-FieldSpec® 3 operates VNIR and SWIR reflectance spectral range. • VNIR and SWIR provide ready-to-use information on the degree of sulphation of carbonate stone. • SWIR data allowed information on efficacy of three different cleaning techniques. • The method supports monitoring of stone alteration during restoration. In urban environments, degradation processes affecting both natural and artificial carbonate materials commonly result in the formation of sulphate-based deposits (i.e., black crusts). Hyperspectral techniques may provide ready-to-use information on the degree of sulphation of carbonate stone, hence helping to monitor the conservation state of carbonate stone monuments. In this study, the Short-Wave InfraRed (SWIR) hyperspectral technique was tested on the world-famous marble-made Loggia di Baccio d’Agnolo in Santa Maria del Fiore Cathedral of Florence (Italy), affected by extensive “black crust” formation. A SWIR method based on a portable spectroradiometer (ASD Fieldspec® 3) was applied to verify the efficacy of different cleaning methods (i.e., laser, chemical and microbial). The interpretation of the SWIR spectra was based on a full-profile approach, in which the calcite and gypsum contributions were evaluated by weighing the best fit simulated contributions of reference compounds. The technique was able to provide in real time semi-quantitative determinations of the amount of gypsum developed on the stone surface, demonstrating to be a fast and handy routine tool during restoration of artworks.