Successive Projections Algorithm Research Articles

Estimation of Canopy Water Content by Integrating Hyperspectral and Thermal Imagery in Winter Wheat Fields

Rapid and accurate estimation of canopy water content (CWC) is important for agricultural water management and food security. Due to the complexity of dynamic changes in water transport during plant growth, estimation of CWC using a single sensor often leads to high uncertainty in the results. Multi-sensor data fusion is one of the solutions to this problem, but suitable spectral preprocessing methods and data fusion methods still need further research. The objectives of this study were to characterize the performance of two varieties at different growth stages under five water stress conditions and screen hyperspectral sensitive spectral bands by using continuous wavelet transform (CWT) and a successive projection algorithm (SPA). Ultimately, the CWC prediction model of winter wheat hyperspectral characteristic bands and thermal imaging information fusion was created using the GRA algorithm. The results showed that canopy temperature parameters and spectral parameters responded significantly to water deficits in winter wheat. Using the CWT-SPA method, a total of 285 hyperspectral feature bands with wavelet decomposition scales ranging from one to eight were selected. The sensitive bands were mainly distributed in the following ranges: 545–561, 746–1348, 1561–1810, and 2122–2430 nm. The GRA algorithm has good multi-source data model fusion capability, and its constructed prediction model based on hyperspectral and thermal image fusion has high accuracy on the canopy water content in winter wheat (R2 = 0.930, RMSE = 5.44%, nRMSE = 7.94%). Compared to the single-feature spectral model (R2 = 0.864, RMSE = 5.92%, nRMSE = 8.63%) and thermal image CWC prediction model (R2 = 0.813, RMSE = 7.22%, nRMSE = 10.49%), the model prediction accuracy based on the GRA algorithm is increased by 7.64% and 13.69%, respectively.

A CARS-SPA-GA Feature Wavelength Selection Method Based on Hyperspectral Imaging with Potato Leaf Disease Classification.

Early blight and ladybug beetle infestation are important factors threatening potato yields. The current research on disease classification using the spectral differences between the healthy and disease-stressed leaves of plants has achieved good progress in a variety of crops, but less research has been conducted on early blight in potato. This paper proposes a CARS-SPA-GA feature selection method. First, the raw spectral data of potato leaves in the visible/near-infrared light region were preprocessed. Then, the feature wavelengths were selected via competitive adaptive reweighted sampling (CARS) and the successive projection algorithm (SPA), respectively. Then, the two sets of wavelengths were reorganized and duplicates were removed, and secondary feature selection was conducted with genetic algorithm (GA). Finally, the feature wavelengths were fed into different classifiers and the parameters were optimized using a real-coded genetic algorithm (RCGA). The experimental results show that the feature wavelengths selected by the CARS-SPA-GA method accounted only for 9% of the full band, and the classification accuracy of the RCGA-optimized support vector machine (SVM) classification model reached 98.366%. These results show that it is feasible to classify early blight and ladybug beetle infestation in potato using visible/near-infrared spectral data, and the CARS-SPA-GA method can substantially improve the accuracy and detection efficiency of potato pest and disease classification.

Dielectric spectroscopy technology combined with machine learning methods for nondestructive detection of protein content in fresh milk.

To quickly achieve nondestructive detection of protein content in fresh milk, this study utilized a network analyzer and an open coaxial probe to analyze the dielectric spectra of milk samples at 100 frequency points within the 2-20GHz range, focusing on the dielectric constant ε' and the dielectric loss factor ε''. Feature variables were extracted from the full dielectric spectra using the successive projections algorithm (SPA), uninformative variables elimination (UVE), and the combined UVE-SPA method. These variables were then used to develop partial least squares regression (PLSR), support vector machine (SVM), decision tree (DT), random forest (RF), and least squares boosting (LSBOOST) models for predicting protein content. The results showed that ε' decreased monotonically with increasing frequency, while ε'' increased monotonically. The UVE-SPA method for feature extraction demonstrated superior performance, with the UVE-SPA-PLSR model being the best for predicting milk protein content, achieving the highest RC 2=0.998 and RP 2=0.989 and the lowest RMSEC=0.019% and RMSEP=0.032%. This study provides a theoretical reference for evaluating milk quality and developing intelligent detection equipment for natural milk.

Predicting water content in engine oil using SPA-ISSA-BP methods  

Contamination of engine oil with water can accelerate oil oxidation and decomposition, resulting in oil degradation. Therefore, the detection and prediction of water content in engine oil is crucial to ensure the long-term safe operation of engines. In this study, near-infrared spectroscopy was used to analyse engine oils with varying water contents. Spectral data were obtained, and various methods, including Savitzky-Golay (SG) smoothing, orthogonal signal correction (OSC), and successive projection algorithm (SPA) were employed to compare the performance differences of principal component regression models. An Improved Sparrow Searches Algorithm (ISSA) was then applied to optimise the Back-Propagation Neural Network (BPNN) for predicting water content in SAE 10W-30 engine oil. The results showed that the performance of the principal component regression model constructed from spectral data after SG smoothing, OSC, and the SPA feature wavelength selection had improved. The BPNN optimised by the improved Sparrow Search Algorithm accelerated the convergence speed of the model and effectively improved the prediction accuracy of the BPNN. The obtained coefficient of determination (R2) was 0.99103, the root mean square error (RMSE) was 2. 2136 10-4, and the mean absolute error (MAE) was 1. 8889×10-4. These results provide an effective method for detecting and predicting the water content in engine oil.

Rapid identification of cod authenticity based on hyperspectral imaging technology

The high economic value of Atlantic cod makes it prone to fraudulent activities in the market, thus achieving rapid and non-destructive identification of its authenticity has practical significance. This study investigated the hyperspectral imaging (HSI) systems with a Vis-NIR (400 – 1000 nm) and SWIR (900 – 1700 nm) spectral range, for determining the authenticity of Atlantic cod fillets in two frozen and thawed sample states. Results found that the model effect of Vis-NIR data was generally better than SWIR data. Random forest (RF) and Linear discriminant analysis (LDA) models of Vis-NIR data achieved 100 % accuracy. Variable screening algorithms of Successive projections algorithm (SPA) and Variable combination population analysis- iteratively retaining informative variables (VCPA-IRIV) maintained 100 % accuracy of the LDA model at VIS-NIR wavebands while simplifying the data operation burden. Overall, this study suggests that HSI is a promising solution for rapid and non-destructive detection of Atlantic cod authenticity.

Research on the Method of Imperfect Wheat Grain Recognition Utilizing Hyperspectral Imaging Technology

As the primary grain crop in China, wheat holds a significant position in the country’s agricultural production, circulation, consumption, and various other aspects. However, the presence of imperfect grains has greatly impacted wheat quality and, subsequently, food security. In order to detect perfect wheat grains and six types of imperfect grains, a method for the fast and non-destructive identification of imperfect wheat grains using hyperspectral images was proposed. The main contents and results are as follows: (1) We collected wheat grain hyperspectral data. Seven types of wheat grain samples, each containing 300 grains, were prepared to construct a hyperspectral imaging system for imperfect wheat grains, and visible near-infrared hyperspectral data from 2100 wheat grains were collected. The Savitzky–Golay algorithm was used to analyze the hyperspectral images of wheat grains, selecting 261 dimensional effective hyperspectral datapoints within the range of 420.61–980.43 nm. (2) The Successive Projections Algorithm was used to reduce the dimensions of the 261 dimensional hyperspectral datapoints, selecting 33 dimensional hyperspectral datapoints. Principal Component Analysis was used to extract the optimal spectral wavelengths, specifically selecting hyperspectral images at 647.57 nm, 591.78 nm, and 568.36 nm to establish the dataset. (3) Particle Swarm Optimization was used to optimize the Support Vector Machines model, Convolutional Neural Network model, and MobileNet V2 model, which were established to recognize seven types of wheat grains. The comprehensive recognition rates were 93.71%, 95.14%, and 97.71%, respectively. The results indicate that a larger model with more parameters may not necessarily yield better performance. The research shows that the MobileNet V2 network model exhibits superior recognition efficiency, and the integration of hyperspectral image technology with the classification model can accurately identify imperfect wheat grains.

Early detection and lesion visualization of pear leaf anthracnose based on multi-source feature fusion of hyperspectral imaging.

Pear anthracnose, caused by Colletotrichum bacteria, is a severe infectious disease that significantly impacts the growth, development, and fruit yield of pear trees. Early detection of pear anthracnose before symptoms manifest is of great importance in preventing its spread and minimizing economic losses. This study utilized hyperspectral imaging (HSI) technology to investigate early detection of pear anthracnose through spectral features, vegetation indices (VIs), and texture features (TFs). Healthy and diseased pear leaves aged 1 to 5 days were selected as subjects for capturing hyperspectral images at various stages of health and disease. Characteristic wavelengths (OWs1 and OWs2) were extracted using the Successive Projection Algorithm (SPA) and Competitive Adaptive Reweighted Sampling (CARS) algorithm. Significant VIs were identified using the Random Forest (RF) algorithm, while effective TFs were derived from the Gray Level Co-occurrence Matrix (GLCM). A classification model for pear leaf early anthracnose disease was constructed by integrating different features using three machine learning algorithms: Support Vector Machine (SVM), Extreme Learning Machine (ELM), and Back Propagation Neural Network (BPNN). The results showed that: the classification identification model constructed based on the feature fusion performed better than that of single feature, with the OWs2-VIs-TFs-BPNN model achieving a highest accuracy of 98.61% in detection and identification of pear leaf early anthracnose disease. Additionally, to intuitively and effectively monitor the progression and severity of anthracnose in pear leaves, the visualization of anthracnose lesions was achieved using Successive Maximum Angle Convex Cone (SMACC) and Spectral Information Divergence (SID) techniques. According to our research results, the fusion of multi-source features based on hyperspectral imaging can be a reliable method to detect early asymptomatic infection of pear leaf anthracnose, and provide scientific theoretical support for early warning and prevention of pear leaf diseases.

Nondestructively Determining Soluble Solids Content of Blueberries Using Reflection Hyperspectral Imaging Technique

Effectively detecting the quality of blueberries is crucial for ensuring that high-quality products are supplied to the fresh market. This study developed a nondestructive method for determining the soluble solids content (SSC) of blueberry fruit by using a near-infrared hyperspectral imaging technique. The reflection hyperspectral images in the 900–1700 nm waveband range were collected from 480 fresh blueberry samples. An image analysis pipeline was developed to extract the spectrums of blueberries from the hyperspectral images. A regression model for quantifying SSC values was successfully established based on the full range of wavebands, achieving the highest RP2 of 0.8655 and the lowest RMSEP value of 0.4431 °Brix. Furthermore, three variable selection methods, namely the Successive Projections Algorithm (SPA), interval PLS (iPLS), and Genetic Algorithm (GA), were utilized to identify the feature wavebands for modeling. The models calibrated from feature wavebands generated an RMSEP of 0.4643 °Brix, 0.4791 °Brix, and 0.4764 °Brix, as well as the RP2 of 0.8507, 0.8397, and 0.8420 for SPA, iPLS, and GA, respectively. Furthermore, a pseudo-color distribution diagram of the SSC values within blueberries was successfully generated based on established models. This study demonstrated a novel approach for blueberry quality detection and inspection by jointly using hyperspectral imaging and machine learning methodologies. It can serve as a valuable reference for the development of grading equipment systems and portable testing devices for fruit quality assurance.

Non-destructive detection of male and female information in ducklings based on near-infrared spectral wavelength selection and deep learning

The technology for sex identification in ducklings can contribute to increased revenue and cost savings in modern duck farming. However, traditional manual identification techniques require high skill levels and involve significant labor intensity. In this study, a non-destructive, user-friendly, and efficient duckling sex identification technique was proposed using near-infrared spectroscopy and deep learning algorithms. Spectral data from 600 groups of newly hatched ducklings were collected. These data were divided into training, testing, and validation sets in a ratio of 7:2:1. The raw spectral data was preprocessed using the Savitzky-Golay convolution derivative method, which was employed for subsequent spectral feature wavelength extraction and modeling. The characteristic wavelengths were extracted using competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), and uninformative variable elimination combined with SPA (UVE-SPA). Conventional machine learning methods − support vector machine (SVM) and different deep learning models, including multilayer perceptron (MLP), mobile network version 2 (MobileNetV2), and residual neural network (ResNet), were studied and compared. The experiment showed that deep learning algorithms outperform traditional spectral analysis models in terms of classification performance. Furthermore, conducting feature wavelength extraction before constructing the classification model could reduce the model’s testing time and even improve its classification performance. Finally, the four models with better classification performance were validated using a validation set, and the combination of MobileNetV2 model UVE-SPA was selected as the optimized model for ducklings’ gender determination, with a classification accuracy of 98.3 % and an average validation time of 1.1 ms. In summary, the detection model established using near-infrared spectroscopy and MobileNetV2 can achieve non-destructive identification of the gender of ducklings. The findings can provide a preliminary research foundation and technical support for the subsequent design of related online intelligent detection systems.

Near-infrared spectroscopy combined with support vector machine for the identification of Tartary buckwheat (Fagopyrum tataricum (L.) Gaertn) adulteration using wavelength selection algorithms

The frequent occurrence of adulterating Tartary buckwheat powder with crop flours in the market necessitates an urgent need for a simple analysis method to ensure the quality of Tartary buckwheat. This study employed near-infrared spectroscopy (NIRS) for the collection of spectral data from Tartary buckwheat samples adulterated with whole wheat, oat, soybean, barley, and sorghum flours. The competitive adaptive reweighted sampling (CARS) and successive projection algorithm (SPA) were deployed to identify informative wavelengths. By integrating support vector machine (SVM) and partial least squares discriminant analysis (PLS-DA), we constructed qualitative models to discern Tartary buckwheat adulteration. The PLS-DA model exhibited prediction accuracies between 89.78 % and 94.22 %, while the mean-centering (MC)-PLS-DA model showcased impressive predictive accuracy of 93.33 %. Notably, the feature-based Autoscales-CARS-CV-SVM model achieved more excellent identification accuracy. These findings exhibit the excellent potential of chemometrics as a powerful tool for detecting food product adulteration.

Predicting leaf nitrogen content of coffee trees using the canopy hyperspectral reflectance feature bands, vegetation index and machine learning

ABSTRACT Leaf nitrogen content (LNC) is an essential indicator of crop nitrogen status. To rapidly and correctly estimate the LNC using hyperspectral remote sensing, the canopy hyperspectral reflectance of coffee trees treated with five levels of nitrogen fertilization in a greenhouse was obtained in this study. Five methods were used for hyperspectral data preprocessing, namely, Savitzky–Golay (SG) smoothing, a combination of SG and standard normal variate transformation (SG-SNV), a combination of SG and first-order derivative (SG-FD), a combination of SG and second-order derivative (SG-SD), and a combination of SG and multiplicative scatter correction (SG-MSC). Feature wavelengths were extracted using variables combination population analysis (VCPA), competitive adaptive reweighted sampling (CARS), and the combination (CARS-SPA) of CARS and successive projections algorithm (SPA). Vegetation indexes (VIs) were constructed and subjected to correlation analysis and variance inflation factor (VIF) analysis. Linear and nonlinear models including partial least squares regression (PLSR), back propagation neural network (BPNN), extreme learning machine (ELM), random forest regression (RFR), and support vector regression (SVR), were adopted to construct LNC retrieval models for coffee trees. The results indicated that SG-MSC could increase the signal-to-noise ratio of hyperspectral data well. The wavelengths selected by CARS-SPA were more relevant to LNC, and combined with ELM resulted in the best performance of LNC prediction (R2 P = 0.901, RMSEP = 0.825 g·kg−1, RPD = 3.229). Ten VIs were obtained through correlation analysis and VIF, and the VIs-based ELM prediction model also performed moderately well (R2 P = 0.814, RMSEP = 1.131 g·kg−1, RPD = 2.354). By comparing the coffee LNC prediction models established by different methods, two coffee LNC inversion models with better prediction accuracy were obtained, which provide a scientific basis for accurate diagnosis of coffee trees LNC, and are of great significance for optimizing the field management.

Screening the characteristic hyperspectral wavelength variables of peanut leaves based on coupled algorithms to predict SPAD value in peanut

ABSTRACT The accurate acquisition and prediction of the chlorophyll content in peanuts can provide reference for scientifically sound planting and management of peanut crops. In this study, peanut canopy leaves analysed during the flowering and pegging period were taken as the research object, and the spectral data collected by a portable hyperspectral spectrometer were used as the data source to study the inversion model of peanut leaf soil and plant analyser development (SPAD). To address the issues of randomness, the need for multiple iterations, and strong subjectivity in threshold selection in the initial variable set of the random frog (RF) algorithm, an improved RF algorithm was developed based on the sparse group lasso tool. By analysing and comparing the results of variable extraction of the improved RF algorithm and six characteristic wavelength screening algorithms, three algorithms were selected for pairwise coupling. These six algorithms were the correlation coefficient method (CC), successive projections algorithm, iteratively retains informative variables (IRIV), uninformative variable elimination (UVE), competitive adaptive reweighed sampling, and a genetic algorithm. The results show that: (1) the results of CC, UVE, and IRIV algorithms are better in a single algorithm experiment; and (2) in the coupled algorithms experiment, CC-UVE, CC-IRIV, and UVE-IRIV methods can effectively reduce dimensions, and the model has a certain level of stability. Among them, the UVE-IRIV-Extreme gradient boosting model had the highest accuracy, R 2 = 0.652, RMSE = 1.867, and the variable compression rate was as high as 97.84%. The research results can provide a theoretical basis for rapid and accurate monitoring of the SPAD values of peanut leaves.

Non-destructive detection of water adulteration level in fresh milk based on combination of dielectric spectrum technology and machine learning method

To rapidly realize the identification of fresh milk for water adulteration and to predict the amount of water adulteration, this study adopts a coaxial probe and a vector network analyzer to analyze the variation laws of dielectric constant ε', dielectric loss factor ε'' and dielectric loss angle tangent tanδ in the range of 2–20 GHz under 100 frequency points. Soft independent modeling of class analogy (SIMCA), Naive Bayes (NB), K-nearest neighbors (KNN), and support vector machine (SVM) models were built based on full dielectric spectra for qualitative testing of adulteration levels in milk. Feature variables of FDS were extracted by using the successive projections algorithm (SPA) and uninformative variables elimination (UVE). Partial least squares regression (PLSR), support vector regression (SVR), and particle swarm optimization least square support vector regression (PSO-LSSVR) models were built for quantitative testing of adulteration levels in milk. The results demonstrate an increasing trend of ε'' and tanδ with increasing frequency and a decreasing trend of ε'. The ε'-SIMCA model achieves the best effect in distinguishing water adulteration in milk, showing accuracy (ACC) of 1, Sensitivity (SNS) of 1, Specificity (SPC) precision (PRE) of 1, and an F1-score (F1) of 1. The tanδ-SPA-PSO-LSSVR model optimally predicts the optimal prediction effect of moisture content of milk, showing an RP2 of 0.994 and an RMSEP of 0.016 %. This study is conducive to building a more comprehensive knowledge system of water adulteration in fresh milk. Its results can provide a theoretical reference in the development of non-destructive detection instruments for natural milk.

Application of terahertz time-domain spectroscopy and chemometrics-based crested porcupine algorithm in identification of different medicinal parts of Angelica sinensis

ObjectiveAngelica sinensis is one of the commonly used Chinese herbal medicine in traditional Chinese medicine clinic, exhibits different pharmacological characteristics due to variations in the content of active ingredients in its head, body, and tail. Therefore, research on the identification methods of different medicinal parts of Angelica sinensis is of great practical significance. Terahertz Time-Domain Spectroscopy (THz-TDS) technology is widely used in the field of nondestructive testing because of its unique electromagnetic wave characteristics. This study explores the feasibility of combining THz-TDS with chemometrics to identify different medicinal parts of Angelica sinensis. MethodsBy comparing the spectral response characteristics of different parts of Angelica sinensis to various optical parameters, the absorption coefficient spectrum in the 0.6–3.0 THz range was selected, and three types of feature extraction algorithms, namely, joint Competitive Adaptive Reweighted Sampling (CARS), Uninformative Variable Elimination (UVE), and Successive Projections Algorithm (SPA), were used to establish the classification models of Extreme Learning Machine (ELM), Random Forest (RF), and Support Vector Machine (SVM) in turn, and optimize the models by using the crown porcupine algorithm (Crested Porcupine Optimizer (CPO) to optimize the model. ResultsThe research results indicate that the CPO optimizer significantly improved the classification accuracy of the models, with the accuracy of the ELM, RF, and SVM models increasing by 4.36%, 1.11%, and 12.22%, respectively. The SPA-CPO-SVM model exhibited the best overall performance, achieving accuracies of 96.11% and 97.96% on the prediction and training sets, respectively, while the number of input features was only 5% of the total feature set. ConclusionThe results show that the fully joint feature extraction strategy and optimization algorithm can play a powerful synergistic effect in model construction, confirming the feasibility of THz-TDS technology to correctly identify different medicinal parts of Angelica sinensis, and providing an important reference for the application of terahertz technology in the identification of Chinese herbal medicines.

Nondestructive detection of multiple qualities of dried jujube in different storage periods based on hyperspectral imaging combined with deep learning

Moisture content and sugar content are characteristic substances that determine the quality of dried jujubes. In this study, we collected visible/near-infrared and near-infrared hyperspectral data (Vis-NIR and NIR) from three dried jujubes samples with different storage periods, and established recognition models for different storage periods of jujubes using Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Deep Learning Methods (LeNet, ResNet, DenseNet, MobileNet and EfficientNet). Then, we constructed a dried jujubes storage period classification model based on feature bands using Successive projections algorithm (SPA) and Principal component analysis (PCA) methods, the results showed that the classification results based on Vis-NIR were better than those based on NIR, and the classification model based on SPA method was more accurate. In addition, the internal quality attributes of moisture content and total sugar during a single storage period of jujube were also predicted, and a quality prediction model for multiple storage periods was established based on NIR. This study provides a fast and non-destructive method for detecting the storage and quality components of jujube, in order to control the quality of jujube at different storage periods.

Non-Destructive Identification of Wool and Cashmere Fibers Based on Cascade Optimizations of Interval-Wavelength Selection Using NIR Spectroscopy

ABSTRACT Near-infrared (NIR) spectroscopy is an effective method for identifying wool and cashmere fibers, with high spectral data providing a wealth of information. However, a key issue is that the accuracy and robustness of subsequent estimates can be reduced by redundant and interfering wavelengths. For this reason, a novel interval-wavelength cascaded optimization method is proposed. Initially, the collected spectral data are preprocessed by standard normal variate transformation (SNV) to eliminate the scattering effect. Then, the backward interval partial least squares (BiPLS) algorithm is applied for the preliminary selection of spectral intervals, followed by the application of three different variable selection algorithms, competitive adaptive reweighted sampling (CARS), successive projection algorithm (SPA) and whale optimization algorithm (WOA), for secondary wavelength optimization, respectively. Finally, both support vector machine (SVM) and random forest (RF) discriminant models are built to identify the extracted subset of wavelengths. In the experimental stage, the cascade method BiPLS-WOA selects 36 wavelengths, in SVM, the accuracy of the validation set reaches 96.9%, and the area under the ROC curve (AUC) can reach 99.3%. The results demonstrate that the proposed method can eliminate redundant and collinear variables, thereby validating the effectiveness of distinguishing wool and cashmere fibers.

Precision variety identification of shelled and in-shell pecans using hyperspectral imaging with machine learning

Pecans are essential nuts containing polyunsaturated fatty acids and dietary fiber which offer health benefits to humans. They are exported and sold in both in-shell and shelled forms. However, varietal identification poses a challenge to both producers and processors, which results in variety substitution for economic advantages. The aim of this study was to investigate the efficacy of visible near-infrared (VNIR) and near-infrared (NIR) hyperspectral imaging systems (HSI) to accurately identify pecan cultivars (Cape fear, Desirable, Stuart blend, and Sumner). The in-shell and shelled spectra were acquired using VNIR and NIR-HSI at wavelengths 400–1000 nm and 900–1700 nm, respectively. The spectra dimensionality was reduced using principal component analysis (PCA). Thereafter, the selected principal components (PCs) were used to build six machine learning classifiers (Decision Tree, Random Forest, Gradient Boosting, Partial Least Square Discriminant Analysis, Support Vector Machine, and Linear Discriminant Analysis (LDA)) for four-class classification. LDA with and without PCA achieved the highest accuracy for both pecan forms. For shelled pecans, the LDA without PCA achieved 90.59 % and increased to 91.67 % accuracy with PCA on the VNIR spectra, while the LDA without PCA achieved 93.36 % and increased to 93.52 % accuracy with PCA on the NIR spectra. For the in-shell pecans, LDA without PCA achieved 98.59 % and increased to 99.12 % accuracy with PCA on the VNIR spectra, while LDA with and without PCA achieved 98.26 % accuracy for the NIR spectra. Moreover, Successive Projection Algorithm was also implemented for wavelength selection and modeling with satisfactory results. Overall, higher accuracy was achieved in the in-shell pecan. This study revealed the usefulness of HSI systems in identifying pecan varieties.

Research on variety identification of common bean seeds based on hyperspectral and deep learning

Accurate, fast and non-destructive identification of varieties of common bean seeds is important for the cultivation and efficient utilization of common beans. This study is based on hyperspectral and deep learning to identify the varieties of common bean seeds non-destructively. In this study, the average spectrum of 3078 hyperspectral images from 500 varieties was obtained after image segmentation and sensitive region extraction, and the Synthetic Minority Over-sampling Technique (SMOTE) was used to achieve the equilibrium of the samples of various varieties. A one-dimensional convolutional neural network model (IResCNN) incorporating Inception module and residual structure was proposed to identify seed varieties, and Support Vector Machine (SVM), K-Nearest Neighbor (KNN), VGG19, AlexNet, ResNet50 were established to compare the identification effect. After analyzing the effects of multiple spectral preprocessing methods on the model, the study selected Savitzky-Golay smoothing correction (SG) for spectral preprocessing and extracted 66 characteristic wavelengths using Successive Projections Algorithm (SPA) as inputs to the discriminative model. Ultimately, the IResCNN model achieved the highest accuracy of 93.06 % on the test set, indicating that hyperspectral technology can accurately identify bean varieties, and the study provides a correct method of thinking for the non-destructive classification of multi-species small-sample bean varieties.

Sentinel-2 Multispectral Satellite Remote Sensing Retrieval of Soil Cu Content Changes at Different pH Levels

With the development of multispectral imaging technology, retrieving soil heavy metal content using multispectral remote sensing images has become possible. However, factors such as soil pH and spectral resolution affect the accuracy of model inversion, leading to low precision. In this study, 242 soil samples were collected from a typical area of the Pearl River Delta, and the Cu content in the soil was detected in the laboratory. Simultaneously, Sentinel-2 remote sensing image data were collected, and two-dimensional and three-dimensional spectral indices were established. Constructing independent decision trees based on pH values, using the Successive Projections Algorithm (SPA) combined with the Boruta algorithm to select the characteristic bands for soil Cu content, and this was combined with Optuna automatic hyperparameter optimization for ensemble learning models to establish a model for estimating Cu content in soil. The research results indicated that in the SPA combined with the Boruta feature selection algorithm, the characteristic spectral indices were mainly concentrated in the spectral transformation forms of TBI2 and TBI4. Full-sample modeling lacked predictive ability, but after classifying the samples based on soil pH value, the R2 of the RF and XGBoost models constructed with the samples with pH values between 5.85 and 7.75 was 0.54 and 0.76, respectively, with corresponding RMSE values of 22.48 and 16.12 and RPD values of 1.51 and 2.11. This study shows that the inversion of soil Cu content under different pH conditions exhibits significant differences, and determining the optimal pH range can effectively improve inversion accuracy. This research provides a reference for further achieving the efficient and accurate remote sensing of heavy metal pollution in agricultural soil.

Estimation Model for Maize Multi-Components Based on Hyperspectral Data.

Assessing the quality of corn seeds necessitates evaluating their water, fat, protein, and starch content. This study integrates hyperspectral imaging technology with chemometric analysis techniques to achieve non-invasive and rapid detection of multiple key components in corn seeds. Hyperspectral images of the embryo surface of maize seeds were collected within the wavelength range of 1100~2498 nm. Subsequently, image segmentation techniques were applied to extract the germ structure of the corn seeds as the region of interest. Seven spectral data preprocessing algorithms were employed, and the Detrending Transformation (DT) algorithm was identified as the optimal preprocessing method through comparative analysis using the Partial Least Squares Regression (PLSR) model. To reduce spectral redundancy and streamline the prediction model, three algorithms were employed for characteristic wavelength extraction: Successive Projections Algorithm (SPA), Competitive Adaptive Reweighted Sampling (CARS), and Uninformative Variable Elimination (UVE). Using the original spectra and extracted characteristic wavelengths, PLSR, BP, RBF, and LSSVM models were constructed to detect the content of four components. The analysis indicated that the CARS-LSSVM algorithm had the best prediction performance. The PSO algorithm was employed to further optimize the parameters of the LSSVM model, thereby improving the model's prediction performance. The R values for the four components in the test set were 0.9884, 0.9490, 0.9864, and 0.9687, respectively. This indicates that hyperspectral technology combined with the DT-CARS-PSO-LSSVM algorithm can effectively detect the main component content of corn seeds. This study not only provides a scientific basis for the evaluation of corn seed quality but also opens up new avenues for the development of non-destructive testing technology in related fields.