Partial Least Squares Regression Model Research Articles

Multivariate regional deep learning prediction of soil properties from near-infrared, mid-infrared and their combined spectra

Artificial neural network (ANN) models have been successfully used in infrared spectroscopy research for the prediction of soil properties. They often show better performance than conventional methods such as partial least squares regression (PLSR). In this paper we develop and evaluate a multivariate extension of ANN for predicting correlated soil properties: total carbon (C), total nitrogen (N), clay, silt, and sand contents, using visible near-infrared (vis-NIR), mid-infrared (MIR) or combined spectra (vis-NIR + MIR). We hypothesize that accounting for the correlation through joint modelling of soil properties with a single model can eliminate “pedological chimera”: unrealistic values that may arise when properties are predicted independently such as when calculating ratio or soil texture values. We tested two types of ANN models, a univariate (ANN-UV) and a multivariate model (ANN-MV), using a dataset of 228 soil samples collected from Murehwa district in Zimbabwe at two soil depth intervals (0–20 and 20–40 cm). The models were compared with results from a univariate PLSR (PLSR-UV) model. We found that the multivariate ANN model was better at conserving the observed correlations between properties and consequently gave realistic soil C:N and C:Clay ratios, but that there was no improvement in prediction accuracy over using a univariate model (ANN or PLSR). The use of combined spectra (vis-NIR + MIR) did not make any significant improvements in prediction accuracy of the multivariate ANN model compared to using the vis-NIR or MIR only. We conclude that the multivariate ANN model is better suited for the prediction of multiple correlated soil properties and that it is flexible and can account for compositional constrains. The multivariate ANN model helps to keep realistic ratio values – with strong implications for assessment studies that make use of such predicted soil values.

Unveiling the transferability of PLSR models for leaf trait estimation: lessons from a comprehensive analysis with a novel global dataset.

Leaf traits are essential for understanding many physiological and ecological processes. Partial least squares regression (PLSR) models with leaf spectroscopy are widely applied for trait estimation, but their transferability across space, time, and plant functional types (PFTs) remains unclear. We compiled a novel dataset of paired leaf traits and spectra, with 47 393 records for > 700 species and eight PFTs at 101 globally distributed locations across multiple seasons. Using this dataset, we conducted an unprecedented comprehensive analysis to assess the transferability of PLSR models in estimating leaf traits. While PLSR models demonstrate commendable performance in predicting chlorophyll content, carotenoid, leaf water, and leaf mass per area prediction within their training data space, their efficacy diminishes when extrapolating to new contexts. Specifically, extrapolating to locations, seasons, and PFTs beyond the training data leads to reduced R2 (0.12-0.49, 0.15-0.42, and 0.25-0.56) and increased NRMSE (3.58-18.24%, 6.27-11.55%, and 7.0-33.12%) compared with nonspatial random cross-validation. The results underscore the importance of incorporating greater spectral diversity in model training to boost its transferability. These findings highlight potential errors in estimating leaf traits across large spatial domains, diverse PFTs, and time due to biased validation schemes, and provide guidance for future field sampling strategies and remote sensing applications.

Prediction of cyanidin 3-rutinoside content in Michelia crassipes based on near-infrared spectroscopic techniques

Currently the determination of cyanidin 3-rutinoside content in plant petals usually requires chemical assays or high performance liquid chromatography (HPLC), which are time-consuming and laborious. In this study, we aimed to develop a low-cost, high-throughput method to predict cyanidin 3-rutinoside content, and developed a cyanidin 3-rutinoside prediction model using near-infrared (NIR) spectroscopy combined with partial least squares regression (PLSR). We collected spectral data from Michelia crassipes (Magnoliaceae) tepals and used five different preprocessing methods and four variable selection algorithms to calibrate the PLSR model to determine the best prediction model. The results showed that (1) the PLSR model built by combining the blockScale (BS) preprocessing method and the Significance multivariate correlation (sMC) algorithm performed the best; (2) The model has a reliable prediction ability, with a coefficient of determination (R2) of 0.72, a root mean square error (RMSE) of 1.04%, and a residual prediction deviation (RPD) of 2.06. The model can be effectively used to predict the cyanidin 3-rutinoside content of the perianth slices of M. crassipes, providing an efficient method for the rapid determination of cyanidin 3-rutinoside content.

Differentiation of mezcales from four agave species using FT-MIR and multivariate statistical analysis

Fourier Transform Mid-Infrared (FT-MIR) spectroscopy and multivariate statistical analysis were used to differentiate mezcales elaborated with four agave species. The FT-MIR data matrix was subjected to spectral transformations using first and second derivatives. The Partial Least Squares (PLS)-Discriminant Analysis (DA) with the matrix transformed by the first and second derivative allowed the differentiation of mezcales. While Orthogonal Partial Least Squares-Discriminant Analysis (OPLS-DA) was more robust when it was analyzed with second-derivative data. Pairwise comparisons by OPLS-DA allowed mezcales to be correctly discriminated, mainly between Agave karwinskii and Agave potatorum (Q2 = 0.654 and p – value < 0.01; R2Y = 0.985 and p-value < 0.01) and between Agave angustifolia and Agave karwinskii (Q2 = 0.563 and p-value = 0.01; R2Y = 0.989 and p-value = 0.01). FT-MIR spectrophotometry and the PLS-Regression (PLS-R) were applied to predict the ethanol percentage (% v/v) of mezcales collected in 2022 based on the PLS-R model previously run on samples evaluated in 2021.

Machine Learning and Feature Selection for Soil Spectroscopy. An Evaluation of Random Forest Wrappers to Predict Soil Organic Matter, Clay, and Carbonates

Soil spectroscopy estimates soil properties using the absorption features in soil spectra. However, modelling soil properties with soil spectroscopy is challenging due to the high dimensionality of spectral data. Feature Selection wrapper methods are promising approaches to reducing dimensionality but are barely used in soil spectroscopy. The aim of this study is to evaluate the performance of two feature selection wrapper methods, Sequential Forward Selection (SFS) and Sequential Flotant Forward Selection (SFFS) built using the Random Forest (RF) algorithm, for dimensionality reduction of spectral data and predictive modelling of modelling soil organic matter (SOM), clay and carbonates. The reflectance of 100 soil samples, acquired from Sierra de las Nieves (Spain), was measured under laboratory conditions using ASD FieldSpec Pro JR. Four different datasets were obtained after applying two spectral preprocessing methods to raw spectra: raw spectra, Continuum Removal (CR), Multiplicative Scatter Correction (MSC), and a so-called “Global” dataset composed of raw, CR and MSC features. The performance of RF models built with feature selection methods was compared to that of Partial Least Squares Regression (PLSR) and RF (alone).RF models built with SFS and SFFS outperformed PLSR and RF alone models: The best RF models with feature selection had a respective ratio of performance to interquartile distance of 1.93, 0.38 and 2.56. PLSR models had an accuracy of 1.41, 0.29 and 1.81 for SOM, carbonates, and clay, respectively. RF alone had a respective performance of 1.29, 0.29 and 1.81. The application of feature selection wrapper methods reduced the number of features to less than 1% of the starting features. Features were selected across all spectra for SOM and clay, and around 900 nm, 1900 nm, and 2350 nm for carbonates. However, feature selection highlighted features around 1100 nm in SOM modelling, as well as other features around 2200 nm, which is considered a main absorption feature of clay. The application of feature selection with Random Forest was very important in improving modelling accuracy, reducing the redundant features and avoiding the curse of dimensionality or Hughes effect. Thus, this research showed an alternative to dimensionality reduction approaches that have been applied to date to model soil properties with spectroscopy and paves the way for further scientific investigation based on feature selection methods and machine learning.

Modeling Risk Assessment of Soil Heavy Metal Pollution using Partial Least Squares and Fuzzy Logic: A Case Study of a Gully Type Coal-Based Solid Waste Dumpsite

Continuous release and migration of heavy metals from coal-based solid waste (CSW) dumpsites often results in significant encroachment on ecological lands and pollution of natural environments. As a result, there is an urgent need for long-term and rapid monitoring, analysis, and assessment to control environmental risks associated with large CSW dumpsites. We constructed a new composite model (PLS-FL) that uses partial least squares regression (PLSR) and fuzzy logic inference (FLI) to accurately predict heavy metal concentrations in soils and assess pollution risk levels. The potential application of the PLS-FL was tested through a gully type CSW case study. We compared 20 modeling strategies using the PLS-FL: five types heavy metals (Cd, Zn, Pb, Cr and As) * four spectral transformation methods (first derivative (FD), second derivative (SD), reverse logarithm (RL), and continuum removal (CR)) * one variable selection method (competitive adaptive reweighted sampling (CARS)). The results showed that the combination of derivative transformation and CARS was recommended for estimation, with R2 C > 0.80 and R2 P > 0.50. When comparing the PLSR model with four traditional machine learning methods (Support Vector Machines (SVM), Random Forests (RF), Extreme Learning Machines (ELM), and KNN), the PLSR model demonstrated the highest average prediction accuracy. Additionally, the FLI process no longer relies on human perception and expert opinion, enhancing the model's objectivity and reliability. The evaluation results revealed that the heavy metal contamination areas of the CSW dumpsite are concentrated at the bottom of the gully, with more severe contamination in the north. Furthermore, a high-risk zone exists in the interim storage area for CSW to the east of the dump. These findings align with the initial detections at the sampling sites and highlight the need for targeted monitoring and control in these areas. The application of the model will empower regulators to quickly assess the overall situation of large-scale heavy metal pollution and provide scientific program and data support for continuous large-scale pollution risk monitoring and sustainable risk management.

Benchtop NMR Coupled with Chemometrics: A Workflow for Unveiling Hidden Drug Ingredients in Honey-Based Supplements.

Recently, benchtop nuclear magnetic resonance (NMR) spectrometers utilizing permanent magnets have emerged as versatile tools with applications across various fields, including food and pharmaceuticals. Their efficacy is further enhanced when coupled with chemometric methods. This study presents an innovative approach to leveraging a compact benchtop NMR spectrometer coupled with chemometrics for screening honey-based food supplements adulterated with active pharmaceutical ingredients. Initially, fifty samples seized by French customs were analyzed using a 60 MHz benchtop spectrometer. The investigation unveiled the presence of tadalafil in 37 samples, sildenafil in 5 samples, and a combination of flibanserin with tadalafil in 1 sample. After conducting comprehensive qualitative and quantitative characterization of the samples, we propose a chemometric workflow to provide an efficient screening of honey samples using the NMR dataset. This pipeline, utilizing partial least squares discriminant analysis (PLS-DA) models, enables the classification of samples as either adulterated or non-adulterated, as well as the identification of the presence of tadalafil or sildenafil. Additionally, PLS regression models are employed to predict the quantitative content of these adulterants. Through blind analysis, this workflow allows for the detection and quantification of adulterants in these honey supplements.

Estimating the Aboveground Fresh Weight of Sugarcane Using Multispectral Images and Light Detection and Ranging (LiDAR)

This study aimed to develop a remote sensing method for estimating the aboveground fresh weight (AGFW) of sugarcane using multispectral images and light detection and ranging (LiDAR). Remotely sensed data were acquired from an unmanned aerial vehicle (drone). Sample plots were harvested and the AGFW of each plot was measured. Sugarcane crown heights and volumes were obtained by isolating individual tree crowns using a LiDAR-derived digital surface model of the area. Multiple linear regression (MLR) and partial least-squares regression (PLSR) models were tested for the field-sampled AGFWs (dependent variable) and individual canopy heights and volumes, and spectral indices were used as independent variables or predictors. The PLSR model showed more promising results than the MLR model when predicting the AGFW over the study area. Although PLSR is well-suited to a large number of collinear predictor variables and a limited number of field samples, this study showed moderate results (R2 = 0.5). The visual appearance of the spatial distribution of the AGFW map is satisfactory. The limited no. of field samples overfitted the MLR prediction results. Overall, this research highlights the potential of integrating remote sensing technologies in the sugarcane industry, thereby improving yield estimation and effective crop management.

Electronic nose for odor monitoring at a landfill fenceline: training and validation of a model for real-time odor concentration measurement

In recent years, electronic noses, or more generally Instrumental Odor Monitoring Systems (IOMS), have aroused increasing interest in the field of environmental monitoring.One of the most interesting applications of these instruments is the real-time estimation of the odor concentration at plant fencelines to continuously monitor odor emissions and identify anomalous conditions. In this type of application, it is possible to setting a “warning” threshold, enabling the continuous check of proper functioning of the plant and sudden intervention in case of malfunctions, preventing, at the same time, the risk of odor events at the receptors. For this purpose, it is necessary to provide a continuous, fast and reliable measurement of the odor concentration, which is nowadays one of the main challenges of this technology.In this context, this work proposes the development of a quantification model for quantifying odors detected at the fenceline of a landfill characterized by very different odor fingerprints. A double-step quantification model, firstly identifying the different odor classes to which the ambient air monitored at the fenceline by the IOMS belong to, and then developing different specific PLS regression models for each of the odor classes identified, was developed. The results of the proposed quantification model were compared to the ones obtained developing a “global” quantification model, which implements the regression on the globality of the training set, without differentiating between the odor classes. Then, they were further evaluated by comparison with the odor events detected at the sensitive receptor by another electronic nose. Moreover, the combined evaluation of the odor events at the plant fenceline and the receptor, respectively, together with the meteorological data highlighted the need of identifying variable warning thresholds for the odor concentrations at the fenceline according to effectively account for meteorological conditions and produce an output that is more correlated with the probability that an odor is perceived outside of the plant.

Assessing the Severity of Verticillium Wilt in Cotton Fields and Constructing Pesticide Application Prescription Maps Using Unmanned Aerial Vehicle (UAV) Multispectral Images

Cotton Verticillium wilt is a common fungal disease during the growth of cotton, leading to the yellowing of leaves, stem dryness, and root rot, severely affecting the yield and quality of cotton. Current monitoring methods for Verticillium wilt mainly rely on manual inspection and field investigation, which are inefficient and costly, and the methods of applying pesticides in cotton fields are singular, with issues of low pesticide efficiency and uneven application. This study aims to combine UAV remote sensing monitoring of cotton Verticillium wilt with the precision spraying characteristics of agricultural drones, to provide a methodological reference for monitoring and precision application of pesticides for cotton diseases. Taking the cotton fields of Shihezi City, Xinjiang as the research subject, high-resolution multispectral images were collected using drones. Simultaneously, 150 sets of field samples with varying degrees of Verticillium wilt were collected through ground data collection, utilizing data analysis methods such as partial least squares regression (PLSR) and neural network models; additionally, a cotton Verticillium wilt monitoring model based on drone remote sensing images was constructed. The results showed that the estimation accuracy of the PLSR and BP neural network models based on EVI, RENDVI, SAVI, MSAVI, and RDVI vegetation indices were 0.778 and 0.817, respectively, with of 0.126 and 0.117, respectively. Based on this, an analysis of the condition of the areas to be treated was performed, combining the operational parameters of agricultural drones, resulting in a prescription map for spraying against cotton Verticillium wilt.

Exploring the Performance of Linear and Nonlinear Models of Time-of-Flight Secondary Ion Mass Spectrometry Spectra.

Multivariate statistical tools and machine learning (ML) techniques can deconvolute hyperspectral data and control the disparity between the number of samples and features in materials science. Nevertheless, the importance of generating sufficient high-quality sample replicates in training data cannot be overlooked, as it fundamentally affects the performance of ML models. Here, we present a quantitative analysis of time-of-flight secondary ion mass spectrometry (ToF-SIMS) spectra of a simple microarray system of two food dyes using partial least-squares (PLS, linear) and random forest (RF, nonlinear) algorithms. This microarray was generated by a high-throughput sample preparation and analysis workflow for fast and efficient acquisition of quality and reproducible spectra via ToF-SIMS. We drew insights from the bias-variance trade-off, investigated the performances of PLS and RF regression models as a function of training data size, and inferred the amount of data needed to construct accurate and reliable regression models. In addition, we found that the spectral concatenation of positive and negative ToF-SIMS spectra improved the model performances. This study provides an empirical basis for future design of high-throughput microarrays and multicomponent systems, for the purpose of analysis with ToF-SIMS and ML.

Estimating Soil Erodible Fraction Using Multivariate Regression and Proximal Sensing Data in Arid Lands, South Egypt

Estimating soil erodible fraction based on basic soil properties in arid lands is a valuable research topic in the field of soil science and land management. The Proximal Sensing (PS) technique offers a non-destructive and efficient method to assess wind erosion potential in arid regions. By using Partial Least Squares Regression (PLSR) and Support Vector Machine (SVM) models and combining soil texture and chemical properties, determined through Visible-Near Infrared (vis-NIR) spectroscopy in 96 soil samples, this study aims to predict soil erodibility, soil organic matter (SOM), and calcium carbonate equivalent (CaCO3) in arid lands located in Elkobaneyya Valley, Aswan Governorate, Egypt. Results showed that the soil erodibility fraction (EF-Factor) had the highest values and possessed a strong relationship between slope and SOM of 0.01% in determining soil erodibility. The PLSR model performed better than SVM for estimating SOM, CaCO3, and EF-Factor. Furthermore, the results showed that the spectral responses of CaCO3 were observed in separate places in the wavelengths of 570, 649, 802, 1161, 1421, 1854, and 2362 nm, and the wavelengths with SOM parameter were 496, 658, 779, 1089, 1417, 1871, and 2423 nm. The EF-factor shows the highest significant correlation with spectral reflectance values at 526, 688, 744, 1418, 1442, 2292, and 2374 nm. The accuracy and performance of the PLSR model in estimating the EF-Factor using spectral reflectance data and the distribution of data points for both the calibration and validation data-sets indicate a good accuracy of the PLSR model, with RMSE values of 0.0921 and 0.0836 Mg h MJ−1 mm−1, coefficient of determination (R2) values of 0.931 and 0.76, and RPD values of 2.168 and 2.147, respectively.

Automated and non-destructive estimation of soluble solid content of tomatoes on the plant under variable light conditions

In this study, a method is proposed to deal with the variable light conditions in a greenhouse to non-destructively predict the soluble solid content (SSC) of tomatoes on the plant. It was investigated how well the SSC – measured as °Brix – of tomatoes could be predicted based on spectral data in the range of 470–900 nm, where data acquired in situ (in the greenhouse) was compared to post-harvest data captured under controlled laboratory conditions. To deal with the variation in illumination in the greenhouse, a dynamic-calibration method is proposed, using a grey reference in the image. Ground-truth SSC data of the tomatoes was acquired using a refractometer. Data was collected of three different types of truss tomatoes with a wide range of SSC. Different PLS regression models were then trained on the spectral data in relation to the refractometer values. Trained and tested on all types, the in situ measurements showed a predicted coefficient of determination on the test set, Q2, of 0.95 with a Root Mean Squared Error of Prediction (RMSEP) of 0.29 °Brix using the dynamic calibration, and a Q2 of 0.93 with RMSEP of 0.35 °Brix without using the dynamic-calibration method. The post-harvest measurements resulted in a Q2 of 0.95 with RMSEP of 0.31 °Brix. The results show that spectral imaging using dynamic calibration is applicable for in situ non-destructive prediction of SSC. This method enables high-throughput and non-destructive quality estimation of fruits on the plant in commercial greenhouse conditions.

A fresh-cut papaya freshness prediction model based on partial least squares regression and support vector machine regression

This study investigated the physicochemical and flavor quality changes in fresh-cut papaya that was stored at 4 °C. Multivariate statistical analysis was used to evaluate the freshness of fresh-cut papaya. Aerobic plate counts were selected as a predictor of freshness of fresh-cut papaya, and a prediction model for freshness was established using partial least squares regression (PLSR), and support vector machine regression (SVMR) algorithms. Freshness of fresh-cut papaya could be well distinguished based on physicochemical and flavor quality analyses. The aerobic plate counts, as a predictor of freshness of fresh-cut papaya, significantly correlated with storage time. The SVMR model had a higher prediction accuracy than the PLSR model. Combining flavor quality with multivariate statistical analysis can be effectively used for evaluating the freshness of fresh-cut papaya.

Rapid estimates of leaf litter chemistry using reflectance spectroscopy

Measuring the chemical traits of leaf litter is important for understanding plants’ roles in nutrient cycles, including through nutrient resorption and litter decomposition, but conventional leaf trait measurements are often destructive and labor-intensive. Here, we develop and evaluate the performance of partial least-squares regression (PLSR) models that use reflectance spectra of intact or ground leaves to estimate leaf litter traits, including carbon and nitrogen concentration, carbon fractions, and leaf mass per area (LMA). Our analyses included more than 300 samples of senesced foliage from 11 species of temperate trees, including needleleaf and broadleaf species. Across all samples, we could predict each trait with moderate-to-high accuracy from both intact-leaf litter spectra (validation R2 = 0.543-0.941; %RMSE = 7.49-18.5) and ground-leaf litter spectra (validation R2 = 0.491-0.946; %RMSE = 7.00-19.5). Notably, intact-leaf spectra yielded better predictions of LMA. Our results support the feasibility of building models to estimate multiple chemical traits from leaf litter of a range of species. In particular, the success of intact-leaf spectral models allows non-destructive trait estimation in a matter of seconds, which could enable researchers to measure the same leaves over time in studies of nutrient resorption.

Near-infrared spectroscopy to quantify overall thermal process intensity during high-moisture extrusion of soy protein concentrate

High-moisture extrusion (HME) is widely used to produce meat analogues. During HME the plant-based materials experience thermal and mechanical stresses. It is complicated to separate their effects on the final products because these effects are interrelated. In this study we hypothesize that the intensity of the thermal treatment can explain a large part of the physicochemical changes that occur during extrusion. For this reason, near–infrared (NIR) spectroscopy was used as a novel method to quantify the thermal process intensity during HME. High–temperature shear cell (HTSC) processing was used to create a partial least squares (PLS) regression curve for processing temperature under controlled processing conditions (root mean standard error of cross-validation (RMSECV) = 4.00 °C, coefficient of determination of cross-validation (R2CV) = 0.97). This PLS regression model was then applied to HME extrudates produced at different screw speeds (200–1200 rpm) and barrel temperatures (100–160 °C) with two different screw profiles to calculate the equivalent shear cell temperature as a measure for thermal process intensity. This equivalent shear cell temperature reflects the effects of changes in local temperature conditions, residence time and thermal stresses. Furthermore, it can be related to the degree of texturization of the extrudates. This information can be used to gain new insights into the effect of various process parameters during HME on the thermal process intensity and extrudate quality.

Enhancing corn quality prediction: Variable selection and explainable AI in spectroscopic analysis

This study addresses the challenge of effectively selecting relevant variables and providing interpretable insights in spectroscopic analysis (1100–2498 nm) of corn quality (moisture, fat, protein, and starch), incorporating variable selection techniques and explainable artificial intelligence (AI). Three variable selection algorithms were used to select 36 important variables, and through combinatorial optimization, only 11 common wavelengths were chosen. Partial least squares regression (PLSR) models were developed using the full spectral range, individual variables, and common variables. Based on the root mean square errors of prediction (RMSEP), the PLSR models with common variables outperformed individual feature variables for starch (0.22 % vs. 0.24 %) and were comparable for oil (0.05 % vs. 0.05 %), water (0.02 % vs. 0.01 %) and protein (0.12 % vs. 0.09 %). The SHapley Additive exPlanation (SHAP) method was employed to explain the PLSR model and assess the contribution of common variables. This combined approach significantly improved the accuracy and interpretability of corn prediction models, providing trust in the results and facilitating the understanding of the relationship between spectral features and corn quality attributes.

Quantitative measurement of internal quality of carrots using hyperspectral imaging and multivariate analysis

The study aimed to measure the carotenoid (Car) and pH contents of carrots using hyperspectral imaging. A total of 300 images were collected using a hyperspectral imaging system, covering 472 wavebands from 400 to 1000 nm. Regions of interest (ROIs) were defined to extract average spectra from the hyperspectral images (HIS). We developed two models: least squares support vector machine (LS-SVM) and partial least squares regression (PLSR) to establish a quantitative analysis between the pigment amounts and spectra. The spectra and pigment contents were predicted and correlated using these models. The selection of EWs for modeling was done using the Successive Projections Algorithm (SPA), regression coefficients (RC) from PLSR models, and LS-SVM. The results demonstrated that hyperspectral imaging could effectively evaluate the internal attributes of carrot cortex and xylem. Moreover, these models accurately predicted the Car and pH contents of the carrot parts. This study provides a valuable approach for variable selection and modeling in hyperspectral imaging studies of carrots.

Tracking moisture contents in the pollution layer on a composite insulator surface using hyperspectral imaging technology.

Electrical insulators used in transmission lines and outdoor substations are exposed to severe environmental pollution, which significantly increases the risk of power system failure, especially when the pollution layer is highly humid due to adverse weather conditions. The focus of this paper is to establish an effective method for assessing the moisture content (MC) in pollution layers as it serves as a crucial indicator for evaluating the risk of failure in insulators. Hyperspectral imaging (HSI) technology with a spectral range of 371.08-1037.89 nm was applied to determine significant changes in reflectance spectral characteristics in insulators during dynamic wetting and drying periods. Partial least squares regression (PLSR) models were utilized to evaluate the data presentation enhancement abilities of spectral transformation models and the data dimensionality reduction abilities of characteristic band selection methods. Furthermore, PLSR models were developed to calculate the MC along the pixel dimension to visually retrieve the dynamic wetting and drying processes of the pollution layer. The R-squared and root-mean-square error (RMSE) results in the cross-verification set and prediction set of the RE-RF(70%)-PLSR model with two characteristic bands with a wavelength of 543.28 nm and 848.01 nm were as follows: RCV2 = 0.9824, RMSECV = 0.0367, RP2 = 0.9818, RMSEP = 0.0369, respectively. This research contributes towards the visualization retrieval of the MC and offers an important technique for analyzing flashover evolution, optimizing insulator design, and preparing coating materials for insulators.

Utilizing the Fusion Characteristics of Multispectral and Electronic Noses to Detect Soil Main Nutrient Content

Soil nutrients are essential for plant growth, and it is crucial to accurately detect their levels. However, current methods for detecting soil nutrients still have low accuracy and high costs. In this study, we propose a multi-source fusion feature approach that combines multispectral image technology with electronic nose gas response technology to achieve rapid, high-precision, and cost-effective detection of soil nutrients, including soil organic matter (SOM), total nitrogen (TN), available potassium (AK), and available phosphorus (AP). To begin, a multispectral camera was used to collect spectral images and extract seven index features. Additionally, an electronic nose was used to collect soil pyrolysis gases to obtain the response curve and extract seven response features. These two sets of features were effectively fused to form a 106 × 98 fusion feature space. Secondly, principal component analysis (PCA) and Pearson correlation coefficient (PCC) algorithms were used to optimize and reduce the dimensionality of the fused feature space. Finally, a predictive model of the relationship between the feature space and the nutrient content of the soil samples was established using the random forest (RF) and partial least squares regression (PLSR) algorithms. The predictive performance of the model was used to evaluate the accuracy of soil nutrient detection. The results showed that the PLSR modeling of the optimized feature space of PCA achieved high accuracy in predicting the levels of main soil nutrients, with R2 values for SOM, TN, AK, and AP of 0.96, 0.95, 0.84, and 0.73, and the RMSE values of 0.56, 0.07, 3.21, and 3.70, respectively. Compared to using only electronic nose gas response technology, the soil nutrient detection with multi-source data features method in this study not only improved the accuracy for SOM and TN but also for metal elements AK and AP in soil. The spectral index features proposed in this study were able to compensate for the limitations of the electronic nose response features, and modeling with the fusion feature space resulted in an accurate prediction of SOM, TN, AK, and AP levels.