Calibration Transfer Methods Research Articles

Transferability of soil organic matter prediction models between different soil types based on improved standard-sample calibration transfer method

Transferability of soil organic matter prediction models between different soil types based on improved standard-sample calibration transfer method

A bidirectional domain separation adversarial network based transfer learning method for near-infrared spectra

Traditional calibration transfer (CT) methods usually fail to adapt the source domain model to the target domain because of changes associated with the instrument, detection environment, sample composition, or sample type. Most deep transfer learning (DTL) methods have shown reliable capability in extracting domain-invariant representations. However, factors unique to the spectral domain limit their development and application, including difficulty in obtaining labeled samples, numerous spectral perturbation factors, and the scarcity of spectral databases. To address these challenges, we here introduce the Bidirectional Domain-Separating Adversarial Network (BiDSAN), based on a deep adversarial network transfer architecture, for learning and adapting to distribution/domain differences between sample spectra. BiDSAN incorporates Wasserstein divergence (w-div) and dynamic adversarial factor into the domain critic (discriminator) to accelerate the extraction of spectral shared components. Additionally, the use of the spectral signal distortion ratio as a loss function and the introduction of a bidirectional mechanism further enhanced the training accuracy and speed of spectral models. Calibration analyses were tested on pharmaceutical and two sets of wood datasets, corresponding to three common domain/distribution difference scenarios discussed in the study. Compared to existing CT and DTL methods, the proposed BiDSAN method demonstrated superior spectral signal calibration capability and stability, achieving unsupervised domain adaptation for similar samples and, for the first time, obtaining ideal prediction results in cross-species spectral domain adaptation. Our method represents significant potential for using DTL approaches in spectral domain adaptation.

Calibration transfer by likelihood maximization: A standard-free method for multiple instruments

Most calibration transfer methods involve transfer between two instruments. Yet, in the context of manufacturing, a model might need to be applied to multiple instruments. Another typical constraint is the acquisition of transfer standards for both instruments with the same variables. These limitations become even more restrictive for multiple probes. Some methods are capable of building models adapted to multiple instruments, but all instruments must be considered during calibration. As such, additional instruments cannot be included a posteriori without changing the initial calibration. Calibration transfer by likelihood maximization has the potential to be applied to multiple instruments without creating a new model each time a new instrument is added. The method is also standard-free and can deal with instruments that have different variables. In this paper, we demonstrate the method’s capacity to use the same model to predict protein content in wheat samples from spectra acquired on four different instruments.

Calibration transfer via filter learning

BackgroundCalibration transfer is an essential activity in analytical chemistry in order to avoid a complete recalibration. Currently, the most popular calibration transfer methods, such as piecewise direct standardization and dynamic orthogonal projection, require a certain amount of standard or reference samples to guarantee their effectiveness. To achieve higher efficiency, it is desirable to perform the transfer with as few reference samples as possible. ResultsTo this end, we propose a new calibration transfer method by using a calibration database from a master instrument (source domain) and only one spectrum with known properties from a slave instrument (target domain). We first generate a counterpart of this spectrum in the source domain by a multivariate Gaussian kernel. Then, we train a filter to make the response function of the slave instrument equivalent to that of the master instrument. To avoid the need for labels from the target domain, we also propose an unsupervised way to implement our method. Compared with several state-of-the-art methods, the results on one simulated dataset and two real-world datasets demonstrate the effectiveness of our method. SignificanceTraditionally, the demand for certain amounts of reference samples during calibration transfer is cumbersome. Our approach, which requires only one reference sample, makes the transfer process simple and fast. In addition, we provide an alternative for performing unsupervised calibration transfer. As such, the proposed method is a promising tool for calibration transfer.

Transfer of near infrared calibration for gasoline octane number based on screening consistent wavelengths combined with direct standardization algorithm

In order to share multivariate calibration models of gasoline research octane number between different near infrared spectrometers, a novel calibration transfer method, namely combination of screening consistent wavelengths and direct standardization (SWCSS-DS) was proposed. Firstly, screening wavelengths with consistent and stable signals (SWCSS) between instruments was used to select the wavelengths with best stability, and then direct standardization (DS) further corrected the systematic errors that still exist after the SWCSS was implemented. The spectra of 120 standard gasoline samples collected on two near infrared spectrometers of the same type were investigated in detail to verify the validity of the new algorithm. Compared results of other transfer methods such as SWCSS, Slope/Bias (S/B), direct standardisation (DS), and piecewise direct standardization (PDS), the root mean squared error for prediction (RMSEP) of SWCSS-DS algorithm for target samples was decreased from 5.75 to 0.295, and the Akaike information criterion value decreased from 1516 to 640, which were better than those of the SWCSS, S/B, DS and PDS algorithms. Therefore, the joint algorithm of SWCSS-DS has not only improved the universality of the master model, but also reduced the dimension of the spectral matrix and calibration equation, that would provide a more efficient model transfer strategy for the practical applications.

Accuracy calibration transfer between multi-sensor systems based on metal-oxide gas sensors used for analysis of exhaled air

The application of electronic nose system as an instrument for exhaled air analysis for diagnosing various diseases (lung cancer, etc.) is one of the most rapidly developing areas of predictive medicine. However, a significant disadvantage of such systems is difficulty in production of identical multi-sensor systems because the sensors of the same type are not identical, that degrades the predictive ability. To solve this problem calibration transfer methods are used.This paper analyzes currently applied calibration transfer methods. A new approach to evaluate the effectiveness of calibration transfer methods for two multi-sensor systems is proposed and implemented. One- and three-component gas mixtures are taken for the analysis. In this work, four methods are used to adjust the sensor responses: UDS, UDSwoi, DS-L1R and DS-PLS2. Applicability of one-factor standardization methods (UDS and UDSwoi) is shown on the classification tasks: multi-class classification task of one-component gas mixtures and binary classification task of three-component gas mixtures. Multi-class classification of model one-component gas mixtures for UDS, UDSwoi with 4 calibration transfer samples provided the classification accuracy of 0.985 and 0.990, respectively. The binary classification of two three-component mixtures with different ratios of three VOCs within the 20% discrepancy limits, UDS, UDSwoi and DS-PLS2 achieved an error-free accuracy of 1.000.

Class-specific correction and classification of NIR spectra of edible oils

Classification of NIR spectra is a common problem in chemometrics. An ideal classification model should not only classify spectra into the known defined classes but also detect spectra that do not belong to any of the known classes. On top of this, the classification model should also perform well on spectra with unwanted source of variation (batch effects) obtained under different conditions. Classification of spectra into more distinct classes (different edible oils) is relatively easier than classification of spectra into less distinct classes (purities of oils). It is crucial to eliminate the batch effects while keeping the signal that separates the classes in the spectra effectively.Classification of spectra obtained under different batch conditions is often performed by applying a classification model (PCLDA, SIMCA) on spectra adjusted using calibration transfer methods (PDS, CTCCA and PCPDS). However, the simple calibration transfer models are unable to successfully remove the batch effects due to the heterogenous nature of these effects. In this paper, we propose CSCAC, a class-specific correction and classification technique that simultaneously corrects batch effects and performs both multi-class classification and novel-class detection. It provides a simple but an effect way to use calibration transfer models class-specifically when the class of the test spectra is unknown/to-be determined. We built three different versions of CSCAC models with PDS, PCPDS and CTCCA as base transfer models and benchmarked against commonly used classification methods. We show that class-specific framework is worth articulating explicitly and existing methods benefit from being used via such a framework. We illustrate its classification performance on a 14-edible oil dataset obtained across 5 different batches. We further illustrate its ability to be a quality check model on peanut oil – maize oil mixtures obtained across 25 batches.

Physicochemical Responsive Integrated Similarity Measure (PRISM) for a Comprehensive Quantitative Perspective of Sample Similarity Dynamically Assessed with NIR Spectra.

Determining sample similarity underlies many foundational principles in analytical chemistry. For example, calibration models are unsuitable to predict outliers. Calibration transfer methods assume a moderate degree of sample and measurement dissimilarities between a calibration set and target prediction samples. Classification approaches link target sample similarities to groups of similar class samples. Although similarity is ubiquitous in analytical chemistry and everyday life, quantifying sample similarity is without a straightforward solution, especially when target domain samples are unlabeled and the only known features are measurable, such as spectra (the focus of this paper). The process proposed to assess sample similarity integrates spectral similarity information with contextual considerations among source analyte contents, model, and analyte predictions. This hybrid approach named the physicochemical responsive integrated similarity measure (PRISM) amplifies hidden-but-essential physicochemical properties encoded within respective spectra. PRISM is tested on four near-infrared (NIR) data sets for four diverse application areas to show efficacy. These applications are the assessment of prediction reliability and model updating for model generalizability, outlier detection, and basic matrix matching evaluation. Discussion is provided on adapting PRISM to classification problems. Results indicate that PRISM collects large amounts of similarity information and effectively integrates it to produce a quantitative similarity evaluation between the target sample and a source domain. The approach is also useful for biological samples with additional physiochemical variations. While PRISM is dynamically tested on NIR data, parts of PRISM were previously applied to other data types, and PRISM should be applicable to other measurement systems perturbed by matrix effects.

Comparison of Transfer Learning and Established Calibration Transfer Methods for Metal Oxide Semiconductor Gas Sensors

Although metal oxide semiconductors are a promising candidate for accurate indoor air quality assessments, multiple drawbacks of the gas sensors prevent their widespread use. Examples include poor selectivity, instability over time, and sensor poisoning. Complex calibration methods and advanced operation modes can solve some of those drawbacks. However, this leads to long calibration times, which are unsuitable for mass production. In recent years, multiple attempts to solve calibration transfer have been made with the help of direct standardization, orthogonal signal correction, and many more methods. Besides those, a new promising approach is transfer learning from deep learning. This article will compare different calibration transfer methods, including direct standardization, piecewise direct standardization, transfer learning for deep learning models, and global model building. The machine learning methods to calibrate the initial models for calibration transfer are feature extraction, selection, and regression (established methods) and a custom convolutional neural network TCOCNN. It is shown that transfer learning can outperform the other calibration transfer methods regarding the root mean squared error, especially if the initial model is built with multiple sensors. It was possible to reduce the number of calibration samples by up to 99.3% (from 10 days to approximately 2 h) and still achieve an RMSE for acetone of around 18 ppb (15 ppb with extended individual calibration) if six different sensors were used for building the initial model. Furthermore, it was shown that the other calibration transfer methods (direct standardization and piecewise direct standardization) also work reasonably well for both machine learning approaches, primarily when multiple sensors are used for the initial model.

PFCE2: A versatile parameter-free calibration enhancement framework for near-infrared spectroscopy

Near-infrared (NIR) spectroscopy is a widely used technique for chemical analysis, but it has faced challenges of calibration transfer, maintenance, and enhancement among different instruments and conditions. The parameter-free calibration enhancement (PFCE) framework was developed to address these challenges with non-supervised (NS), semi-supervised (SS), and full-supervised (FS) methods. This study presented PFCE2, an updated version of the PFCE framework that incorporates two new constraints and a new method to improve the robustness and efficiency of calibration enhancement. First, normalized L2 and L1 constraints were introduced to replace the correlation coefficient (Corr) constraint used in the original PFCE. These constraints preserve the parameter-free feature of PFCE and impose smoothness or sparsity on the model coefficients. Second, multitask PFCE (MT-PFCE) was proposed within the framework to address the calibration enhancement among multiple instruments, enabling the framework to be versatile for all possible calibration transfer situations. Demonstrations conducted on three NIR datasets of tablets, plant leaves, and corn showed that the PFCE methods with the new L2 and L1 constraints can result in more accurate and robust predictions than the Corr constraint, especially when the standard sample size is small. Moreover, MT-PFCE could refine all models in the involved scenarios at once, leading to significant enhancement in model performance, compared to the original PFCE method with the same data requirements. Finally, the applicable situations of the PFCE framework and other analogous calibration transfer methods were summarized, facilitating users to choose suitable methods for their application. The source codes written in both MATLAB and Python are available at https://github.com/JinZhangLab/PFCE and https://pypi.org/project/pynir/, respectively.

Review: The evolution of chemometrics coupled with near infrared spectroscopy for fruit quality evaluation. II. The rise of convolutional neural networks

The Part 1 prequel to this review evaluated the evolution of modelling techniques used in evaluation of fruit quality over the past three decades and noted a progression towards the use of artificial neural networks (ANNs) and convolutional neural networks (CNNs). In this review, Part 2, the use of CNNs for NIR fruit quality evaluation is explored, given the success of CNNs in various other fields, such as image, video, speech, and audio processing, and the availability of large (open source) datasets of fruit spectra and reference quality attribute, which is required for the training of CNN models. The review provides an overview of deep learning and the CNN architectures and techniques used in NIR spectroscopy for regression modelling, with advantages and disadvantages identified. Studies using CNN for NIR based fruit quality evaluation are then critically examined. Eight publications have presented on models using the same open-source mango dry matter calibration and test set, enabling inter-method comparisons. CNN models have been demonstrated to be accurate, precise and robust. Techniques of transfer learning for CNN models offer an alternative solution to model updating and calibration transfer methods applied in traditional chemometrics. The review has highlighted crucial areas that require resolution and exploration in this application through future research, including, (i) data requirements for training a CNN (ii) optimal spectral pre-processing for CNN (iii) CNN architecture and hyper-parameter selection and tuning for fruit quality evaluation (iv) CNN model interpretability and explainability. Future studies must conduct clearer comparison to partial least squares (PLS) regression and shallow ANNs to better assess the prospective benefit of using CNN, a more complex model. The potential for visualisation of spectra relevance to the CNN model using techniques such as GradCam, currently employed in visualising 2D-CNN models, remains to be explored.

Calibration transfer for near-infrared (NIR) spectroscopy based on local preserving projection

Calibration transfer is a mathematical or statistical technique to solve the universality of near-infrared (NIR) spectroscopy, which is particularly important in practical applications. Therefore, a calibration transfer method based on local preserving projection (CTLPP) to correct spectral differences was proposed in this paper. Local preserving projection (LPP) is a linear feature extraction algorithm, which is a linear approximation of Laplacian Eigenmap manifold learning. It can not only overcome the shortcomings of linear feature extraction algorithms which are difficult to maintain the nonlinear manifold of the original data, but also solve the problem that nonlinear manifold learning methods are difficult to directly map new samples, which is conducive to building the conversion relationship between the source spectra and the target spectra. The transfer performance of the proposed method was evaluated in three datasets, and compared with five approaches, namely, piecewise direct standardization (PDS), calibration transfer based on independent component analysis (CTICA), spectral space transformation (SST), principal components canonical correlation analysis (PC-CCA) and calibration transfer based on neighborhood preserving embedding (CTNPE). Experimental results indicated that this proposed method can successfully correct the detection spectra among different apparatus. In addition, it can obtain relatively low prediction root mean square error (RMSEP). Therefore, the comprehensive research conducted in this work demonstrated that the proposed method is a promising calibration transfer method in NIR applications, which can provide a reference when sharing the NIR spectral models with more and more instruments.

Novel Hybrid Calibration Transfer Method Based on Nonlinear Dimensionality Reduction for Robust Standardization in Near-Infrared Spectroscopy

To correct the spectral changes measured on different instruments in near-infrared (NIR) spectroscopy, a novel hybrid calibration transfer method with nonlinear dimensionality reduction based on direct standardization (DS) and autoencoder (AE), named DS-AE, is reported. First, DS was employed to preliminarily eliminate the spectral difference from the master and slave instruments. Next, the spectral features were extracted by AE to construct the partial least squares (PLS) calibration model. Compared with the linear dimensionality reduction methods, AE learns more latent features of the input spectra that are beneficial to reflect the chemical information of samples. Two NIR experiment datasets, including wheat and corn samples measured on different spectrometers, were employed to evaluate the performance of the DS-AE method. DS, piecewise direct standardization (PDS), canonical correlation analysis (CCA), principal components canonical correlation analysis (PC-CCA), and transfer via an extreme learning machine auto-encoder (TEAM) were introduced for comparative analysis with the proposed method. The results showed that DS-AE provided the lowest root mean squared error of prediction (RMSEP) for the wheat dataset; For the corn dataset, DS-AE provided lower RMSEP than DS, PDS, and CCA, and comparable with PC-CCA and TEAM. The score of the PLS principal components (PCs) describes the spectral differences of different instruments. The results indicated that the hybrid DS-AE method effectively corrected for the spectral variations. In summary, the proposed hybrid DS-AE method provided an alternative for robust standardization of near-infrared spectra measured on different instruments.

Calibration transfer by likelihood maximization: A standard-free approach capable of handling non-overlapping wavelength ranges

Calibration transfer is useful when modeling data collected from different, yet similar, sources. This may occur when using spectrophotometers to monitor an industrial process. The objective of this method is to compensate for differences between the instruments and reduce any biases that may arise. Calibration transfer seeks to apply a multivariate model to a validation dataset that differs in some meaningful way from the calibration set. However, making robust calibration transfer models typically requires the user to acquire many observations in both datasets and ensure that some observations are present in both the calibration and validation datasets. Our novel methodology seeks to address these issues, which are often cumbersome in an industrial context. The proposed methodology is not a typical calibration transfer method but has the same role as it estimates the observations acquired from one X2 domain into another X1 domain. We implement a bootstrapping strategy to quantify the estimates of the spectra in the new domain and the uncertainty associated with them. These results are then combined with an inverse multivariate regression to predict the y values, which is known as the likelihood maximization inverse regression (LMIR) methodology. In this paper, we first present the algorithms used for estimation and prediction. The proposed method is tested on two real datasets and the results are compared with those of three other methods that are typically used to perform direct calibration, namely, partial least squares (PLS), powered PLS and direct LMIR. A comparison to calibration methods is made here to provide a reference for the results obtained and demonstrate the benefits of using our estimation model. These results reveal that the proposed methodology outperforms the standard predictive methods. Further, it is beneficial because it does not need standard samples and is capable of handling different wavelength ranges and variables.

Improved Principal Component Analysis (IPCA): A Novel Method for Quantitative Calibration Transfer between Different Near-Infrared Spectrometers.

Given the labor-consuming nature of model establishment, model transfer has become a considerable topic in the study of near-infrared (NIR) spectroscopy. Recently, many new algorithms have been proposed for the model transfer of spectra collected by the same types of instruments under different situations. However, in a practical scenario, we need to deal with model transfer between different types of instruments. To expand model applicability, we must develop a method that could transfer spectra acquired from different types of NIR spectrometers with different wavenumbers or absorbance. Therefore, in our study, we propose a new methodology based on improved principal component analysis (IPCA) for calibration transfer between different types of spectrometers. We adopted three datasets for method evaluation, including public pharmaceutical tablets (dataset 1), corn data (dataset 2), and the spectra of eight batches of samples acquired from the plasma ethanol precipitation process collected by FT-NIR and MicroNIR spectrometers (dataset 3). In the calibration transfer for public datasets, IPCA displayed comparable results with the classical calibration transfer method using piecewise direct standardization (PDS), indicating its obvious ability to transfer spectra collected from the same types of instruments. However, in the calibration transfer for dataset 3, our proposed IPCA method achieved a successful bi-transfer between the spectra acquired from the benchtop and micro-instruments with/without wavelength region selection. Furthermore, our proposed method enabled improvements in prediction ability rather than the degradation of the models built with original micro spectra. Therefore, our proposed method has no limitations on the spectrum for model transfer between different types of NIR instruments, thus allowing a wide application range, which could provide a supporting technology for the practical application of NIR spectroscopy.

Performance Optimization of a Developed Near-Infrared Spectrometer Using Calibration Transfer with a Variety of Transfer Samples for Geographical Origin Identification of Coffee Beans

This research aimed to improve the classification performance of a developed near-infrared (NIR) spectrometer when applied to the geographical origin identification of coffee bean samples. The modification was based on the utilization of a collection of spectral databases from several different agricultural samples, including corn, red beans, mung beans, black beans, soybeans, green and roasted coffee, adzuki beans, and paddy and white rice. These databases were established using a reference NIR instrument and the piecewise direct standardization (PDS) calibration transfer method. To evaluate the suitability of the transfer samples, the Davies-Bouldin index (DBI) was calculated. The outcomes that resulted in low DBI values were likely to produce better classification rates. The classification of coffee origins was based on the use of a supervised self-organizing map (SSOM). Without the spectral modification, SSOM classification using the developed NIR instrument resulted in predictive ability (% PA), model stability (% MS), and correctly classified instances (% CC) values of 61%, 58%, and 64%, respectively. After the transformation process was completed with the corn, red bean, mung bean, white rice, and green coffee NIR spectral data, the predictive performance of the SSOM models was found to have improved (67-79% CC). The best classification performance was observed with the use of corn, producing improved % PA, % MS, and % CC values at 71%, 67%, and 79%, respectively.

Instant quantification of sugars in milk tablets using near-infrared spectroscopy and chemometric tools

Milk tablets are a popular dairy product in many Asian countries. This research aimed to develop an instant and rapid method for determining sucrose and lactose contents in milk tablets using near-infrared (NIR) spectroscopy. For the quantitative analysis, a training set composed of laboratory-scale milk samples was generated based on a central composite design (CCD) and used to establish partial least squares (PLS) regression for the predictions of sucrose and lactose contents resulting in R2 values of 0.9749 and 0.9987 with the corresponding root mean square error of calibration (RMSEC) values of 1.69 and 0.35. However, the physical difference between the laboratory-scale powder and the final product milk tablet samples resulted in spectral deviations that dramatically affected the predictive performance of the PLS models. Therefore, calibration transfer methods called direct standardization (DS) and piecewise direct standardization (PDS) were used to adjust the NIR spectra from the real milk tablet samples before the quantitative prediction. Using high-performance liquid chromatography (HPLC) as a reference method, the developed NIR-chemometric model could be used to instantly predict the sugar contents in real milk tablets by producing root mean square error of prediction (RMSEP) values for sucrose and lactose of 5.04 and 4.22 with Q2 values of 0.7973 and 0.9411, respectively, after the PDS transformation.

Prediction approach of larch wood density from visible-near-infrared spectroscopy based on parameter calibrating and transfer learning.

Wood density, as a key indicator to measure wood properties, is of weighty significance in enhancing wood utilization and modifying wood properties in sustainable forest management. Visible–near-infrared (Vis-NIR) spectroscopy provides a feasible and efficient solution for obtaining wood density by the advantages of its efficiency and non-destructiveness. However, the spectral responses are different in wood products with different moisture content conditions, and changes in external factors may cause the regression model to fail. Although some calibration transfer methods and convolutional neural network (CNN)-based deep transfer learning methods have been proposed, the generalization ability and prediction accuracy of the models still need to be improved. For the prediction problem of Vis-NIR wood density in different moisture contents, a deep transfer learning hybrid method with automatic calibration capability (Resnet1D-SVR-TrAdaBoost.R2) was proposed in this study. The disadvantage of overfitting was avoided when CNN processes small sample data, which considered the complex exterior factors in actual production to enhance feature extraction and migration between samples. Density prediction of the method was performed on a larch dataset with different moisture content conditions, and the hybrid method was found to achieve the best prediction results under the calibration samples with different target domain calibration samples and moisture contents, and the performance of models was better than that of the traditional calibration transfer and migration learning methods. In particular, the hybrid model has achieved an improvement of about 0.1 in both R 2 and root mean square error (RMSE) values compared to the support vector regression model transferred by piecewise direct standardization method (SVR+PDS), which has the best performance among traditional calibration methods. To further ascertain the generalizability of the hybrid model, the model was validated with samples collected from mixed moisture contents as the target domain. Various experiments demonstrated that the Resnet1D-SVR-TrAdaBoost.R2 model could predict larch wood density with a high generalization ability and accuracy effectively but was computation consuming. It showed the potential to be extended to predict other metrics of wood.

Adaptive Artificial Neural Network in near infrared spectroscopy for standard-free calibration transfer

Near infrared spectroscopy (NIRS) is a secondary analytical method that could use machine learning to predict components of interest for various applications. Calibration transfer (CT) algorithms have been used to transfer the existing NIRS calibration model to another NIRS instruments to reduce the involvement of expensive and tedious conventional chemical analytical methods. However, most of CT algorithms involve both primary and secondary instruments to acquire NIRS signals from standard samples for spectral standardization. On the other hand, artificial neural network (ANN) that can adapt to new environmental conditions may be an alternative standard-free CT algorithm in transferring a trained ANN to secondary instruments without the involvement of the primary instrument. Although ANN has been widely implemented in NIRS research as a calibration model, its feasibility as an alternative CT method has not been evaluated. Thus, this study evaluates the feasibility of an adaptive ANN (AANN) in transferring ANN model from primary to secondary instruments using two traceable NIRS datasets. First, ANN and ANN with principal components (PCs-ANN) were developed and optimized using Bayesian learning algorithm. After that, these ANNs were adapted to secondary instruments using respective transfer data, in which, only the weights and biases of the ANNs were updated. Findings show that the lowest averaged RMSEP was achieved by the proposed PCs-AANN and AANN in the corn and wheat datasets, respectively. Particularly, the computation cost of having a calibration model in secondary instruments has been substantially reduced by means of the proposed AANN algorithm.

Model robustness in estimation of blueberry SSC using NIRS

In near-infrared (NIR) spectroscopy analysis of blueberry often the differences between biological variability factors (cultivar, season, etc.) lead to variations in light propagation properties and cell structure. Therefore, non-invasive detection models for blueberry developed based on NIR spectroscopy are often limited and unstable. To accommodate diverse prediction scenarios and improve the accuracy of NIR models in blueberry SSC estimation, a combined model calibration strategy that integrates Global-modeling method and calibration transfer method is proposed in this study. The Global-modeling method and calibration transfer methods (DOP and SBC) were applied for datasets containing only seasonal or cultivar differences and for those containing both seasonal and cultivar differences, respectively. Compared with the prediction model without correction, both Global-modeling and calibration transfer strategies improved model performance in the face of seasonal/cultivar variation challenges, with higher Rp and lower RMSEP values. After the application of Global-modeling, DOP and SBC methods, for the correction of seasonal variables (Bluecrop-2014), the RMSEP values of the model were reduced by 55.9%, 44.7% and 44.5%; for the correction of cultivar variables (M2-2015), the RMSEP values of the model were reduced by 45.8%, 22.8% and 37.9%; for the correction of cultivar variables (Duke-2015), the RMSEP values of the model were reduced by 9.3%, 5.9% and 2.8%. The experimental results indicated that combining Global-modeling and calibration transfer methods can weaken the influence of external conditions on blueberry NIR spectra to a certain extent, enhance the robustness of the model to biological variation, and improve the detection accuracy of blueberry SSC.