Regression Neural Network Research Articles

Revolutionizing Facial Recognition: A Dolphin Glowworm Hybrid Approach for Masked and Unmasked Scenarios

Machine learning has several essential applications, including classification and recognition. Both people and objects may be identified using the Machine learning technique. It is particularly important in the verification process since it recognizes the characteristics of human eyes, fingerprints, and facial patterns. With the advanced technology developments, nowadays, Facial recognition is used as one of the authentication processes by utilizing machine learning and deep learning algorithms and it has been the subject of several academic studies. These algorithms performed well on faces without masks, but not well on faces with masks. since the masks obscured the preponderance of the facial features. As a result, an improved algorithm for facial identification with and without masks is required. After the Covid-19 breakout, deep learning algorithms were utilized in research to recognize faces wearing masks. Those algorithms, however, were trained on both mask- and mask-free faces. Hence, in this, the cropped region for the faces is only used for facial recognition. Here, the features were extracted using the texture features, and the best-optimized features from the glow worm optimization algorithm are used in this paper. With these features set, the hybrid Dolphin glow worm optimization is used for finding the optimal features and spread function value for the neural network. The regression neural network is trained with the optimized feature set and spread function for the face recognition task. The performance of the suggested method will be compared to that of known approaches such as CNN-GSO and CNN for face recognition with and without masks using accuracy, sensitivity, and specificity will next be examined.

Python-based machine learning estimation ofthermo-hydraulic performance along varying nanoparticle shape, nanofluid and tube configuration

In this research article, a Python-based machine learning model prediction study was conducted based on the study results obtained from sudden expansion tubes containing different expansion angles, dimpled fin structures and nanofluids, whose thermo-hydraulic performance was previously examined. In the study, Artificial Neural Network and Ridge regression models were used to make predictions on the average Nusselt number (Nu), average Darcy friction factor (f) and performance evaluation criteria (PEC). Physical variations of the sudden expansion tube were taken into account and a detailed comparison of the results was made. A superior average Nu was acquired as 172.45 %, 22.05 %, 17.18 %, 13.65 %, and 7.76 % compared to Ag-MgO/H2O, Al2O3/H2O (blade), CoFe2O4/H2O, Al2O3/H2O (cylindrical), and Al2O3/H2O (platelet), respectively. The highest Performance Evaluation Criteria (PEC) for Re= 2000 based on Al2O3/H2O (platelet) shows an increase of 4.84 %, 12.08 %, 11.76 %, 66.05 %, and 148.94 % compared to Al2O3/H2O (cylindrical), Al2O3/H2O (blade), CoFe2O4/H2O, Fe3O4/H2O, and Ag-MgO/H2O, respectively. From the results obtained, it was determined that Python-based Machine Learning approach which facilitates custom optimizations showed a significant performance with small margins of error in predicting the heat transfer parameters. The lowest error rates of machine learning and polynomial ridge regression models ranged from 0.2 % to 5.4 % for the unseen test set and the application of Python-based algorithms provided considerable savings in calculation time compared to conventional methods. On the other hand, using machine learning models with feature engineering has been found to increase model performance by at least 30 %. In these years when studies on the predictions of thermo-hydraulic studies are very rare in the literature, this study is intended to facilitate scientists, engineers and academicians who will further study on this subject.

Rate coefficients for C and O2 reactive collisions relevant to interstellar clouds from QCT and machine learning.

The chemical reactions between certain interstellar molecules are exothermic in nature and barrierless in the entrance channel, allowing these reactions to occur rapidly even at low astronomical temperatures, e.g., C and O2 interaction. Obtaining detailed rovibrational transition parameters for the reaction between C and O2, such as state-selected rate coefficients, is crucial for studying the associated atmospheric and astronomical environments. Hence, this work presents an approach that combines quasi-classical trajectory calculations with machine learning techniques based on Neural Network (NN) and Gaussian Process Regression (GPR) to determine state-selected rate coefficients. Employing this approach, we significantly reduced the computational requirements while simultaneously obtaining the accurate state-selected reaction cross sections and rate coefficients for the collision of C and O2. Both the NN-based and GPR-based models established in this work accurately predict the results calculated from explicit numerical calculations in the explored temperature range of 50-1500K, achieving a coefficient of determination R2 > 0.96. Most importantly, the current work provides the most comprehensive dataset of rovibrational rate coefficients of v = 0-4, j = 0-70 → v' = 0-15 for the astrophysical modeling of the C-O2 collision system.

An ultra-fast MMC-HVDC fault location algorithm based on transient voltage features and regression neural network

An ultra-fast fault location algorithm based on the single-ended transient voltage features and regression neural network (RNN) is proposed, which utilizes 2.5 ms postfualt data window and is suitable for modular multilevel converter-based high-voltage DC (MMC-HVDC) grids equipped with quick-action protections and hybrid DC circuit breakers (HDCCBs). Firstly, the analyses based on the lumped RLC equivalent circuit demonstrate that the delay time, the first negative peak time and its value all have exact relationships with the fault location. Nevertheless, considering the actual parameters and topology of the MMC-HVDC grid, three features can only be approximately extracted. Thus, RNN is utilized to estimate fault locations. 2.02 × 104 distinct fault cases validate the algorithm’s high accuracy across all fault locations and transition resistances up to 1005 Ω. It can well tolerate reasonable deviations of line parameter and current limiting reactor value, as well as 40-dB white noise. Besides, it also has remarkable adaptability, and can be suitable for different systems.

Reconstructing urban vegetation evolution in China using multimodal deep learning and 30-years Landsat archive

To retrospect the impact of urbanization on vegetation, consistent and reliable maps that provide urban vegetation estimates are required. In this study, we make the first attempt to employ deep learning for urban vegetation mapping of 2120 cities in China using a 30-years Landsat archive (UV-30). We present a multimodal deep learning (MDL) model to combine features from 5 categories, namely reflective bands, remote sensing indices, texture variables, topographical variables, and time labels. We normalized elevation data locally to remove the inter-city differences, while retaining topographic details. We used time labels to account for spectral variability over time and Landsat sensors. The performance of the MDL models was compared with a vanilla deep neural network, random forest, and support vector regression. Results indicated that the fusion of different categories of features improved the prediction accuracy. The MDL performed best in most validation cities, and our results were superior to the fractional maps aggregated from high-resolution datasets in homogeneous urban landscapes. We further analyzed urban vegetation dynamics across the urban core-new town-outskirt gradients of different geographical regions and city sizes. We found a gradually diminishing de-greening trend in urban China, especially in urban cores and new towns. Nevertheless, we observed a more pronounced degradation of urban vegetation in the south regions of China compared to the north, and smaller cities show no evident urban vegetation increase. The suggested framework is valuable for urban vegetation assessments and facilitates a thorough understanding and management of urban ecosystems.

Performance of Artificial Neural Network Compared to Multi-Linear Regression in Prediction of Countermovement Jump Height

Previous research has used primarily linear regression models to predict jump height and establish contributors of performance. The purpose of this study was to compare the performance of artificial neural network (ANN) and multi-linear regression (MLR) in prediction of countermovement jump (CMJ) height and investigating the contribution of kinematic variables to CMJ performance. Thirty-four healthy young male athletes performed a total of 204 CMJ while eight kinematic variables (the hip, knee, and ankle angles at the begging of the concentric phase of CMJ, the hip and knee take-off angles, and the shoulder, hip, and knee maximum angular velocities) were used as inputs to ANN and MLR to predict CMJ height. The correlation coefficients between the jump height and the predicted value by the developed models indicated that ANN predict CMJ height better than MLR (R2 = 0.68 compared to R2 = 0.44). Moreover, the root mean squared error of prediction showed better performance of the ANN rather than the MLR (4.8 cm compared to 5.3 cm). The shoulder and hip maximum angular velocities were the most important contributors, and then the hip and knee take-off angles contributed to CMJ height. In conclusion, implementing ANN to identify key variables of performance may also be relevant for other sport skills.

Modeling Groundwater Levels in California's Central Valley by Hierarchical Gaussian Process and Neural Network Regression

AbstractModeling groundwater levels continuously across California's Central Valley (CV) hydrological system is challenging due to low‐quality well data which is sparsely and noisily sampled across time and space. The lack of consistent well data makes it difficult to evaluate the impact of 2017 and 2019 wet years on CV groundwater following a severe drought during 2012–2015. A novel machine learning method is formulated for modeling groundwater levels by learning from a 3D lithological texture model of the CV aquifer. The proposed formulation performs multivariate regression by combining Gaussian processes (GP) and deep neural networks (DNN). The hierarchical modeling approach constitutes training the DNN to learn a lithologically informed latent space where non‐parametric regression with GP is performed. We demonstrate the efficacy of GP‐DNN regression for modeling non‐stationary features in the well data with fast and reliable uncertainty quantification, as validated to be statistically consistent with the empirical data distribution from 90 blind wells across CV. We show how the model predictions may be used to supplement hydrological understanding of aquifer responses in basins with irregular well data. Our results indicate that on average the 2017 and 2019 wet years in California were largely ineffective in replenishing the groundwater loss caused during previous drought years.

Prediction of municipal waste generation using multi-expression programming for circular economy: a data-driven approach.

The existing surge in municipal waste generation (MWG), characterized by swiftly changing and uncontrollable factors, poses a significant challenge to sustainable development. This prompted the need for improved predictive models to guide strategic waste management within the circular economy framework. This study aims to develop a predictive model using multi-expression programming (MEP) to assess MWG. The model was developed using historical data on socioeconomic and environmental factors and validated via comparative analyses with artificial neural network (ANN), random forest (RF), and multiple linear regression (MLR) using various evaluation metrics. The parametric and sensitivity analyses of the MEP model were also conducted. The MEP, ANN, RF, and MLR models have a coefficient of determination (R2) (for testing datasets) of 0.977, 0.974, 0.957, and 0.964, respectively. The MEP model is superior in terms of accuracy and performance for the prediction of MWG when compared to the other three models. The sensitivity analysis revealed the relative importance of each input variable in the established MEP model. The novelty of this research lies in the application of MEP to predict MWG and the formulation of a new mathematical model that links socioeconomic and environmental factors with MWG. The model can be used by waste management authorities to optimize waste collection, transportation, and disposal infrastructure for an effective circular economy and sustainable development. This model also aids in the development of effective waste management policies.

Breath Analysis Using Quartz Tuning Forks for Predicting Blood Glucose Levels Using Artificial Neural Networks.

Diabetes Mellitus (DM), a widespread metabolic disorder, poses lifelong health implications, demanding timely diagnosis and cautious monitoring for effective disease management. Traditional blood glucose tests are invasive and require medical expertise for intermittent checking, motivating the investigation of alternative, noninvasive methods. This study introduces an approach employing breath analysis through a set of 12 quartz tuning fork-based sensors enhanced using nanomaterials and dedicated artificial neural network (ANN) algorithms for data interpretation. The breath analysis methodology involves capturing unique breath signatures using the frequency-based sensor array. The accompanying neural network classification algorithm, customized for the sensor data, enables precise classification of data from 245 individuals as diabetic, prediabetic, or healthy. A neural network regression algorithm predicted blood glucose values and was compared with the actual values obtained from medical blood glucose measurement. The clinical relevance of the predicted blood glucose has been examined using error grids. The sensor array coupled with the ANN algorithm can identify diabetic, prediabetic, and control samples with 97% test accuracy. Blood glucose was predicted using neural network regression with a correlation coefficient of 0.89 and a mean square error of 0.13.

Reconstructing a Fine Resolution Landscape of Annual Gross Primary Product (1895–2013) with Tree-Ring Indices

Low carbon management and policies should refer to local long-term inter-annual carbon uptake. However, most previous research has only focused on the quantity and spatial distribution of gross primary product (GPP) for the past 50 years because most satellite launches, the main GPP data source, were no earlier than 1980. We identified a close relationship between the tree-ring index (TRI) and vegetation carbon dioxide uptake (as measured by GPP) and then developed a nested TRI-GPP model to reconstruct spatially explicit GPP values since 1895 from seven tree-ring chronologies. The model performance in both phases was acceptable: We chose general regression neural network regression and random forest regression in Phase 1 (1895–1937) and Phase 2 (1938–1985). With the simulated and real GPP maps, we observed that the GPP for grassland and overall GPP were increasing. The GPP landscape patterns were stable, but in recent years, the GPP’s increasing rate surpassed any other period in the past 130 years. The main local climate driver was the Palmer Drought Severity Index (PDSI), and GPP had a significant positive correlation with PDSI in the growing season (June, July, and August). With the GPP maps derived from the nested TRI-GPP model, we can create fine-scale GPP maps to understand vegetation change and carbon uptake over the past century.

Entrepreneurial firm creation and economic uncertainty: an explainable artificial intelligence approach

ABSTRACT Through this paper, we examined the relationship between macroeconomic factors and new entrepreneurial firms in France during the 2000–2020 period; this relationship included several uncertainties and shocks (the 9/11, the 2008 global financial crisis, the Brexit, COVID-19 Pandemic, and so forth). To this aim, we applied various advanced machine learning (ML) techniques, including the eXtreme Gradient Boosting (XGBoost) algorithm, Light Gradient-Boosting Machine (LightGBM) algorithm, Neural Network (NN), Random Forest (RF), and Linear Regression (LR). We also investigated SHapley Additive exPlanation (SHAP) framework for interpretation and analysis purposes. Furthermore, we compared the performance and forecasting accuracy of the five models. The study revealed three major results. First, the XGBoost algorithm provides the most accurate prediction of a firm’s creation. Second, a heterogeneous relationship exists between new business creation and the business cycle. Third, higher GDP and unemployment levels are associated with greater procyclicality and countercyclicality in entrepreneurship, respectively. Finally, the results of the study suggest the importance of developing effective financial assistance during economic downturns to encourage faster recovery in the formation of new businesses.

Simulation and optimization of cheese whey additive for value-added compost production: Hyperparameter tuning approach and genetic algorithm

Cheese whey is a difficult and costly wastewater to treat due to its high organic matter and mineral content. Although many management strategies are conducted for whey removal, its use in composting is limited. In this study, the effect of cheese whey in the composting of sewage sludge and poultry waste on compost quality and process efficiency was investigated. Also, valid and consistent simulations were developed with Gaussian Process Regression (GPR), Support Vector Regression (SVR), and Neural Network Regression (NNR) Machine Learning (ML) algorithms. The results of all physicochemical parameters determined that 3% of cheese whey addition for both feedstocks improved the composting process's efficiency and the final product's quality. The best results obtained through hyperparameter tuning showed that Gaussian Process Regression (GPR) was the most effective modeling tool providing realistic simulations. The reliability of these simulations was verified by running the GPR process 50 times. MdAPE demonstrated the validity and consistency of the created process simulations. Moreover, a genetic algorithm was used to optimize these dependent simulations and achieved almost 100% desirability. Optimization studies showed that the effective cheese whey ratios were 3.2724% and 3.1543% for sewage sludge and poultry waste, respectively. Optimization results were compatible with the results of experimental studies. This study provides a new strategy for the recovery of cheese whey as well as a new perspective on the effect of cheese whey on both physicochemical parameters and composting phases and the modeling and optimization processes of the results.

Multi-objective optimisation and verification of creep-resistant Ni-base superalloy for electron-beam powder-bed-fusion

This paper reports the use of integrated computational alloy design, coupled with a rapid printability screening method, to downselect from a total of 70000 datasets in design space to five candidates in the first step, and then from five to one in the second step. The new Ni-base superalloy with compositions of Ni-5.03Al-2.69Co-5.63Cr-0.04Hf-1.91Mo-2.36Re-3.32Ta-0.57Ti-8.46W-0.05C-0.019B exhibits an optimal balance of density (8.82 g/cm2), printability (freezing range of 107 °C), thermal stability (γ′-volume fraction of 50.7% at 980 °C and low Md‾ value) and creep (rupture time of 612 h at 980 °C/120 MPa). The micro-hardness varies mildly from 417.2 ± 18.5 to 434.7 ± 14.6 HV, suggesting good phase stability. This is substantiated by microstructure observations, which revealed the absence of a topologically close-packed phase. Machine-learning tools of the artificial neural network (ANN), random forest, and support vector regression, respectively, were used to predict creep rupture time. The ANN algorithm achieves the highest accuracy in predicting creep life. By recognising the “black box” nature of the ANN, interpretability analysis was conducted using the local interpretable model-agnostic method. The analysis supports that the ANN model truly learned meaningful functional relationships, and thus is judged as reliable. Feature correlation evaluation outcome emphasises the importance of incorporating microstructure-related input features.

Electromagnetic Interference Effect Assessment Under Measuring Testability Limitation Based on Physics-Informed Neural Network and Gaussian Process Regression

Electromagnetic Interference Effect Assessment Under Measuring Testability Limitation Based on Physics-Informed Neural Network and Gaussian Process Regression

Non-Destructive Detection of Tea Polyphenols in Fu Brick Tea Based on Hyperspectral Imaging and Improved PKO-SVR Method

Tea polyphenols (TPs) are a critical indicator for evaluating the quality of tea leaves and are esteemed for their beneficial effects. The non-destructive detection of this component is essential for enhancing precise control in tea production and improving product quality. This study developed an enhanced PKO-SVR (support vector regression based on the Pied Kingfisher Optimization Algorithm) model for rapidly and accurately detecting tea polyphenol content in Fu brick tea using hyperspectral reflectance data. During this experiment, chemical analysis determined the tea polyphenol content, while hyperspectral imaging captured the spectral data. Data preprocessing techniques were applied to reduce noise interference and improve the prediction model. Additionally, several other models, including K-nearest neighbor (KNN) regression, neural network regression (BP), support vector regression based on the sparrow algorithm (SSA-SVR), and support vector regression based on particle swarm optimization (PSO-SVR), were established for comparison. The experiment results demonstrated that the improved PKO-SVR model excelled in predicting the polyphenol content of Fu brick tea (R2 = 0.9152, RMSE = 0.5876, RPD = 3.4345 for the test set) and also exhibited a faster convergence rate. Therefore, the hyperspectral data combined with the PKO-SVR algorithm presented in this study proved effective for evaluating Fu brick tea’s polyphenol content.

Groundwater quality prediction and risk assessment in Kerala, India: A machine-learning approach

Despite its critical importance for health, agriculture, and the economy, and its key role in supporting climate change adaptation, groundwater quality remains vulnerable to contamination and is often neglected until significant deterioration. The groundwater resources of Kerala, one of the southernmost states of India, are under escalating stress and scarcity, despite a high well density with 62% of the population relying on groundwater from approximately 6.5 million open wells. This study investigates the detailed hydrogeochemistry and predicts groundwater quality zones of the state using machine-learning techniques viz, extreme gradient boosting (XGBoost), support vector regression (SVR), artificial neural network (ANN) and random forest (RF) regression. The hydrogeochemical analysis reveals varying groundwater quality across the state. Among the different machine learning models, RF shows higher goodness of fit (R2: 0.922) with minimal prediction error (root mean square error: 6.29 and mean absolute error: 3.12). The predicted groundwater quality was validated using the spatially distributed stiff diagrams, visually representing water composition trends of each well. The very good, good, moderate and poor groundwater quality zones occupy 31.7%, 40.4%, 20.4%, and 7.4% of the state aligning accurately with the groundwater quality scenario of the state. Additionally, groundwater drinking risk assessment was conducted, considering that 7.4% of the state experiences poor-quality groundwater. Integrating groundwater quality maps with population data, the study assessed potential health risks due to consuming untreated water. Nearly 0.59 million people across 252 local self-government bodies (LSGs) are susceptible to consuming poor quality groundwater, which may pose potential health risks. This observation provides valuable insights for sustainable groundwater management and public health safeguarding and the findings of the present study are useful for achieving sustainable development goal (SGD) 6 (clean water and sanitation) and long-term groundwater management in Kerala.

Prediction of Potato Volume by Neural Network Regression Model

Potatoes are essential for meeting the world's dietary requirements due to their high concentration of carbohydrates, vitamins, and minerals. Potato production is plentiful worldwide, but postharvest processing losses have reduced potato quality. Potatoes can be graded manually, which is time-consuming and labor-intensive. In this study, a neural network regression model was developed to predict the volume of potatoes using 2D images, which can be used as a parameter for grading potatoes automatically. The image processing method was used to extract features like the major axis, minor axis, and roundness of 400 potato samples, and each potato's volume was thoroughly measured using the water displacement method, and a dataset was generated. These attributes were used to create the neural network regression model and predict the volume. The results revealed that the mean squared error (MSE) was 83.93, the root mean squared error (RMSE) was 9.16, and R2 was approximately 0.882. The potatoes were categorized into three groups: small, medium, and large, based on their size. The utilized model demonstrated an overall accuracy rate of 91.4%. Specifically, it achieved 90% and 85% accuracy for small-sized and medium-sized potatoes, respectively, whereas the model displayed a decrease in accuracy from 60% to 70% for large-sized potatoes. To summarize, the suggested neural network regression model provided an efficient approach to volume prediction using 2D images to improve the grading process of agricultural products.

Predicting the higher heating value of products through solid yield in torrefaction process

Torrefaction is a widely employed method to upgrade biomass for energy purpose. The higher heating value (HHV) serves as a vital indicator for assessing the energy potential of biomass. Nevertheless, HHV measurement is a time-consuming and costly process. HHV of raw biomass, torrefied biomass, and biochar has been extensively estimated using the results of elemental analysis and/or proximate analysis. However, these data must be repeatedly measured for each target object. To reduce the efforts required for the measurement of each object, this study proposes, for the first time, the use of solid yield (the most easily obtainable data) to predict the HHV of torrefied biomass.A total of 215 sets of data were retrieved from different biomass types undergoing various torrefaction technologies. Artificial neural network (ANN) and hierarchical linear regression (HLR) were employed to develop prediction models. Solid yield demonstrated a clear correlation to HHV for each individual biomass; however, diverse biomass types exhibited significant variation. Therefore, the results of elemental analysis of raw feedstocks were incorporated to account for variations in feedstock types and to iterate the models. Generally, ANN is superior to HLR in predicting the HHV of torrefied biomass with the optimized model achieving an R2 of 0.9133. Additionally, two biomasses subjected to two torrefaction technologies were used to validate the model, and an R2 of 0.9676 was achieved.As a result, in a torrefaction process, once the raw biomass is analyzed, the HHV of any torrefied biomass can be predicted simply by measuring the weight of the product.

Rapid Detection of Protein and Starch Content in Brewing Wheat Using Hyperspectral Imaging Technology Combined with a Convolutional Neural Network Regression Model

Wheat is the main raw material in liquor brewing. However, the protein content (PC) and starch content (SC) in wheat will affect the quality and flavor of the final liquor. In this study, the rapid, non-destructive determination of the PC and SC in wheat was achieved by combining hyperspectral imaging (HSI) with a convolutional neural network regression (CNNR) model established using the original spectral data in the hyperspectral images. The best preprocessing method was first determined, and then the performance of CNNR, extreme gradient boosting (XGBoost) and partial least squares regression (PLSR) models were compared at full wavelength, and it was concluded that CNNR based on the original spectrum was the optimal model for predicting wheat PC (R-square (R2) = 0.9942, root mean square error (RMSE) = 0.1041, relative percentage difference (RPD) = 13.1306) and SC (R2 = 0.9329, RMSE = 0.8633, RPD = 3.8605) at full wavelength. Finally, to further explore the feature extraction capability of the CNN model, different feature selection and extraction methods are used to build the PLSR model, and the comparison revealed that the PLSR model built by feature extraction using convolutional neural network (CNN_F) performed best in predicting wheat PC and SC. These results showed that HSI combined with a CNNR model could enable the rapid and accurate analysis of the PC and SC in wheat, since the CNNR can extract features from samples better than the traditional machine learning models.

Inversion Modeling of Chlorophyll Fluorescence Parameters in Cotton Canopy via Moisture Data and Spectral Analysis

The study of chlorophyll fluorescence parameters is very important for understanding plant photosynthesis. Monitoring cotton chlorophyll fluorescence parameters via spectral technology can aid in understanding the photosynthesis, growth, and stress of cotton fields in real time and provide support for cotton growth regulation and planting management. In this study, cotton plot experiments with different water treatments were set up to obtain the spectral reflectance of the cotton canopy, the maximum photochemical quantum yield (Fv/Fm), and the photochemical quenching coefficient (qP) of leaves at different growth stages. Support vector machine regression (SVR), random forest regression (RFR), and artificial neural network regression (ANNR) were used to establish a fluorescence parameter inversion model of the cotton canopy leaves. The results show that the original spectrum was transformed by multivariate scattering correction (MSC), the standard normal variable (SNV), and continuous wavelet transform (CWT), and the model constructed with Fv/Fm passed accuracy verification. The SNV-SVR model at the budding stage, the MSC-SVR model at the early flowering stage, the SNV-SVR model at the full flowering stage, the MSC-SVR model at the flowering stage, and the CWT-SVR model at the full boll stage had the highest estimation accuracy. The accuracies of the three spectral preprocessing and qP models were verified, and the MSC-SVR model at the budding stage, SNV-SVR model at the early flowering stage, MSC-SVR model at the full flowering stage, SNV-SVR model at the flowering stage, and CWT-SVR model at the full boll stage presented the highest estimation accuracies.