Search

• Rapid determination of rice protein content was completed using NIR spectroscopy. • BiPLS-GSA and BiPLS-SABPSO was proposed to select feature wavelengths of protein. • BiPLS-SABPSO was superior to BiPLS-GSA in feature wavelengths selection. • Detection accuracy of regression model was efficiently improved using BiPLS-SABPSO. Protein content is a key indicator for rice nutritional value, and its accurate determination has become a necessary tool for rice quality evaluation and breeding. To solve the shortcomings of slow and costly testing by traditional methods, a rapid detection model of rice protein content was constructed based on near-infrared spectroscopy coupled with feature wavelength selection. The backward interval partial least squares (BiPLS) was combined with the genetic simulated annealing algorithm (GSA) and the simulated annealing binary particle swarm optimization algorithm (SABPSO), respectively, to construct BiPLS-GSA and BiPLS-SABPSO for selecting protein feature wavelengths, thereby establishing a corresponding partial least squares quantitative calibration model. Among them, the regression model established using feature wavelengths selected by BiPLS-SABPSO had the best performance. The determination coefficient of the model for the validation set and the independent test set were 0.949 and 0.956, the root mean square error were 0.174% and 0.214%, the relative root mean square error were 2.621% and 3.118%, and the residual predictive deviation being 4.235 and 4.500, respectively, which could meet the requirement for rapid and accurate determination of rice protein content. The combination of BiPLS-SABPSO and near-infrared spectroscopy has developed into a new method to achieve fast and reliable determination of rice protein content, providing an alternative strategy for rapid qualitative detection of related agricultural products. [ABSTRACT FROM AUTHOR]

• Rapid detection of protein content in rice was performed by Raman-NIR fusion data. • IBPSO was proposed to select spectral characteristic wavelengths of rice protein. • IBPSO was superiority to other four algorithms in Raman-NIR wavelengths selection. • Detection accuracy of Raman-NIR fusion data regression model was improved by IBPSO. Protein content is an essential index for evaluating rice quality. This work discussed the feasibility of rapid detection of protein content in rice using spectral data fusion technology. An improved binary particle swarm optimization algorithm (IBPSO) was proposed to select the characteristic wavelength of Raman and near-infrared spectroscopy fusion data, which improved the detection accuracy of the partial least squares correction model. The determination coefficient of prediction, root mean square error of prediction, and mean relative error of prediction of the protein content detection model established by IBPSO were 0.903, 0.235%, and 2.768%, respectively, which were better than the modeling performance of the other four algorithms. The research shows that IBPSO can efficiently acquire high correlation modeling wavelength variables through the guiding optimization of binary bits with a value of '1′. The combination of IBPSO and spectral data fusion strategy can realize the rapid detection of protein content in rice, which provides theoretical support for developing related online detection equipment. [ABSTRACT FROM AUTHOR]

• A class of networks is designed to generate complete spectral data. • The reliability and usability of generated spectral samples are demonstrated. • Validating superior performance across generated datasets. With the rise of deep learning, the combination of spectroscopy analysis techniques and deep learning methods has been extensively utilized in the field of agriculture, such as the detection of crop diseases, soil analysis, and crop quality assessment. Unfortunately, it is often difficult to obtain enough spectral samples that can be used for deep learning because of environmental constraints, equipment limitations, and labour costs. To address this issue, we propose a spectral sample augmentation technique based on the K-condition boundary equilibrium generative adversarial networks (KC-BEGAN). First, a stable boundary equilibrium generative adversarial network (BEGAN) model is constructed, and the KC-BEGAN model is built by incorporating chemical property labels, multiscale gradient information, and the k-nearest neighbour algorithm. The goal is to enrich complete spectral samples with chemical properties. Second, we compare the differences between generated samples and real samples using methods such as t-distributed stochastic neighbour embedding, Mahalanobis distance, F test, and maximum mean discrepancy. Following data augmentation by the KC-BEGAN model, the R2 on the test set for traditional regression models (PLSR, SVR, RR, PCR) and deep regression models (Inception-ResNet, Inception, 1D-CNN) improved by 4.9%, 0.4%, 2.5%, 4.4%, 3.6%, 4.4%, and 6.8% respectively. Furthermore, this study replaces the GAN module in the KC-BEGAN model with Diffusion-GAN and conducts experiments following the same procedure to evaluate the feasibility of other GAN models in augmenting labelled near-infrared spectral samples. The research results indicate that the spectral samples generated by the KC-BEGAN model are reliable and can meet the expansion requirements of small-scale spectral sample sets. Simultaneously, replacing the GAN module in the KC-BEGAN model with other GAN models is also feasible, suggesting that the continuous label marking method proposed in this study is indeed effective. [ABSTRACT FROM AUTHOR]

The contents of cellulose and hemicellulose (C and H) in corn stover (CS) have an important influence on its biochemical transformation and utilization. To rapidly detect the C and H contents in CS by near-infrared spectroscopy (NIRS), the characteristic wavelength selection algorithms of backward partial least squares (BIPLS), competitive adaptive reweighted sampling (CARS), BIPLS combined with CARS, BIPLS combined with a genetic simulated annealing algorithm (GSA), and CARS combined with a GSA were used to select the wavelength variables (WVs) for C and H, and the corresponding regression correction models were established. The results showed that five wavelength selection algorithms could effectively eliminate irrelevant redundant WVs, and their modeling performance was significantly superior to that of the full spectrum. Through comparison and analysis, it was found that CARS combined with GSA had the best comprehensive performance; the predictive root mean squared errors of the C and H regression model were 0.786% and 0.893%, and the residual predictive deviations were 3.815 and 12.435, respectively. The wavelength selection algorithm could effectively improve the accuracy of the quantitative analysis of C and H contents in CS by NIRS, providing theoretical support for the research and development of related online detection equipment. [ABSTRACT FROM AUTHOR]

Stacking ensemble learning is one of the most effective integration technologies and is increasingly applied to near-infrared spectroscopy combined with chemometrics methods. The prediction accuracy of Stacking is primarily affected by the selection of different models. However, many current studies are mainly artificial selection models' combinations. It affects the model's prediction accuracy and increases the algorithm's difficulty. It is difficult to efficiently and accurately find the optimal configuration scheme. This study applies a genetic algorithm to find the optimal base and meta learner combinations in Stacking ensemble learning. This method uses the near-infrared spectral data set of corn seed germination rate. First, select the best pretreatment methods for seven models, including Gaussian process regression (GPR), SVR, PLS, etc. The above seven single learners after pretreatment are taken as the candidate base learner, and then random forest (RF), SVR, PLS, and GPR are taken as the potential meta learner; use a genetic algorithm to select the optimal model combination configuration and generate GA-Stacking algorithm. The model prediction results of the improved model GA-Stacking are compared with several single models and Stacking ensemble learning via the artificial selection model combinations. The results show that the prediction performance using the GA-Stacking ensemble learning model is optimal, R2 is 0.9022, and RMSE is 0.1100. The experiment shows that the model combination selected by the genetic algorithm has significantly improved the prediction performance of the Stacking ensemble learning model and reduced the risk of the model's overfitting. • Genetic algorithm is used to optimize model combinations for Stacking ensemble learning. • Overcome the disadvantage of artificial selection of stacking ensemble learning model combination optimization. • Verify the effectiveness of the Stacking ensemble learning optimization algorithm. [ABSTRACT FROM AUTHOR]

Near-infrared spectroscopy (NIR) detection technology enables nondestructive analysis of samples. To achieve fast detection of maize seed germination rates using NIR spectroscopy, 245 samples of maize seeds from three different varieties were purchased from the market and subjected to artificial aging to obtain maize seeds with varying levels of vigor. NIR spectroscopy was collected for each sample, and germination experiments were performed. The Monte Carlo cross-validation method was employed to eliminate 12 outliers, and the baseline correction method along with multiplicative scatter correction (MSC) was applied to the NIR. Partial least squares (PLS), support vector machine (SVM) models, and Gaussian process regression (GPR) models based on different kernel functions were established for detecting maize germination rates. Through comparison of the results of different models, it was found that the GPR model based on the Matern32 kernel function performed best. The R² values for the training and testing sets of this model were 0.9727 and 0.9599, respectively. Experimental results showed that the Matern32-GPR model exhibited greater stability and superior capability in detecting maize seed germination rates. This method can provide a reference for other applications based on NIR detection. [Display omitted] [ABSTRACT FROM AUTHOR]

• Biochemical methane potential was rapidly estimated in anaerobic co-digestion. • Mean relative errors of models based on physicochemical indexes were under 3.1%. • Mean relative error of model based on near infrared spectroscopy was 2.333%. • Model based on near-infrared spectroscopy had the highest predicted accuracy. To rapidly estimate the biochemical methane potential (BMP) of feedstocks, different multivariate regression models were established between BMP and the physicochemical indexes or near-infrared spectroscopy (NIRS). Mixed fermentation feedstocks of corn stover and livestock manure were rapidly detected BMP in anaerobic co-digestion (co-AD). The results showed that the predicted accuracy of NIRS model based on characteristic wavelengths selected by multiple competitive adaptive reweighted sampling outperformed all regression models based on the physicochemical indexes. For the NIRS regression model, coefficient of determination, root mean squares error, relative root mean squares error, mean relative error and residual predictive deviation of the validation set were 0.982, 6.599, 2.713%, 2.333% and 7.605. The results reveal that the predicted accuracy of NIRS model is very high, and meet the requirements of rapid prediction of BMP for co-AD feedstocks in practical biogas engineering. [ABSTRACT FROM AUTHOR]

• Biochemical methane potential was rapidly estimated in anaerobic co-digestion. • Mean relative errors of models based on physicochemical indexes were under 3.1%. • Mean relative error of model based on near infrared spectroscopy was 2.333%. • Model based on near-infrared spectroscopy had the highest predicted accuracy. To rapidly estimate the biochemical methane potential (BMP) of feedstocks, different multivariate regression models were established between BMP and the physicochemical indexes or near-infrared spectroscopy (NIRS). Mixed fermentation feedstocks of corn stover and livestock manure were rapidly detected BMP in anaerobic co-digestion (co-AD). The results showed that the predicted accuracy of NIRS model based on characteristic wavelengths selected by multiple competitive adaptive reweighted sampling outperformed all regression models based on the physicochemical indexes. For the NIRS regression model, coefficient of determination, root mean squares error, relative root mean squares error, mean relative error and residual predictive deviation of the validation set were 0.982, 6.599, 2.713%, 2.333% and 7.605. The results reveal that the predicted accuracy of NIRS model is very high, and meet the requirements of rapid prediction of BMP for co-AD feedstocks in practical biogas engineering. [ABSTRACT FROM AUTHOR]