Savitzky-Golay Convolution Research Articles

A Data-Driven Model for Classification of Traditional Chinese Medicine Materials

This paper proposes a data-driven classification model for traditional Chinese medicinal herbs based on mid-infrared spectral data. Addressing the limitations of traditional identification methods when the herbs’ appearance is damaged or incomplete, this study employs machine learning techniques to achieve accurate classification of medicinal herb types through spectral data preprocessing and feature extraction. Firstly, the Savitzky-Golay convolution smoothing method and Standard Normal Variate (SNV) transformation were used for denoising the spectral data. Then, Principal Component Analysis (PCA) was employed to reduce the dimensionality of the high-dimensional spectral data and extract the key features. Finally, the Gaussian Mixture Model (GMM) was applied to cluster the reduced data, categorizing the medicinal herbs into six classes. The results show that this method produces the accurate and stable classification. The constructed model is not only applicable to the classification and origin identification of medicinal herbs but also provides an important reference value for the classification and origin identification of other plant species.

Feasibility and potential of terahertz spectral and imaging technology for Apple Valsa canker detection: A preliminary investigation

Apple Valsa canker (AVC) caused by the Ascomycete Valsa mali, seriously constrains the production and quality of apple fruits. The symptomless incubation characteristics of Valsa mali make it highly challenging to detect AVC at an early infection stage. After infecting the wound of apple bark, the pathogenic hyphae of AVC will expand and colonize the phloem tissue. Meanwhile, various enzymes and toxic substances released by hyphae cause the decomposition of cellulose and lignin, and the generation of poisonous secondary metabolites in bark tissue. However, these early symptoms of AVC are invisible from the bark’s appearance. Fortunately, Terahertz Spectral Imaging (ThzSI) technology with the advantage of penetrating, and fingerprinting is promising for detecting hidden or slight symptoms of the fungal infection. This study is a preliminary investigation of terahertz frequency-domain spectra for AVC in the early stage of infection. Healthy and two-week-infected apple tree branches were prepared for capturing ThzS images, and the spectral data were preprocessed by Multivariate scattering correction (MSC), Savitzky-Golay convolution smoothing (SG), and standard normal variate (SNV) respectively to remove data noise and improve data quality. Principal component analysis (PCA), competitive adaptive reweighted sampling (CARS), and random frog (RFROG) were employed to extract the spectral feature bands to eliminate redundant data and improve computational efficiency. Machine learning models were established based on the spectral features to detect AVC at an early infection stage, where 11 of them exhibited the best performance with F1-score of 99.72%. To further explore disease information in spatial spectra, imaging data were acquired using terahertz imaging technology. Based on imaging data, pseudo-color imaging, histogram equalization, and Otsu segmentation were employed to visualize early infection areas in apple barks. Furthermore, histogram feature (HF), shape feature (SF), and local binary pattern (LBP) extracted from terahertz spectral images were utilized to establish the SVM, RF, and KNN models. HF-SF-KNN and HF-SF-LBP-KNN with the best performance achieved F1-score of 98.82%. This study presents a preliminary application of terahertz spectral and imaging technology for early-stage AVC detection and demonstrates its feasibility. Additionally, it provides a new way to detect AVC, which expands the application of ThzSI technology in tree disease detection in orchards and lays the foundation for further research.

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.

Enhanced prediction of cement raw meal oxides by near-infrared spectroscopy using machine learning combined with chemometric techniques.

The component analysis of raw meal is critical to the quality of cement. In recent years, near-infrared (NIR) has been emerged as an innovative and efficient analytical method to determine the oxide content of cement raw meal. This study aims to utilize NIR spectroscopy combined with machine learning and chemometrics to improve the prediction of oxide content in cement raw meal. The Savitzky-Golay convolution smoothing method is applied to eliminate noise interference for the analysis of calcium carbonate ( ), silicon dioxide ( ), aluminum oxide ( ), and ferric oxide ( ) in cement raw materials. Different wavelength selection techniques are used to perform a comprehensive analysis of the model, comparing the performance of several wavelength selection techniques. The back-propagation neural network regression model based on particle swarm optimization algorithm was also applied to optimize the extracted and screened feature wavelengths, and the model prediction performance was checked and evaluated using and RMSE. In conclusion, the results indicate that NIR spectroscopy in combination with ML and chemometrics has great potential to effectively improve the prediction performance of oxide content in raw materials and highlight the importance of modeling and wavelength selection techniques. By enabling more accurate and efficient determination of oxide content in raw materials, NIR spectroscopy coupled with meta-modeling has the potential to revolutionize quality assurance practices in cement manufacturing.

Modeling of soluble solid content of PE-packaged blueberries based on near-infrared spectroscopy with back propagation neural network and partial least squares (BP-PLS) algorithm.

Blueberries are a nutritious and popular berry worldwide. The physical and chemical properties of blueberries constantly change through the cycle of the supply chain (from harvest to sale). The purpose of this study was to develop a rapid method for detecting the properties of packaged blueberries based on near-infrared (NIR) spectroscopy. NIR was applied to quantitatively determine the soluble solid content (SSC) of polyethylene (PE)-packaged blueberries. An orthogonal partial least squares discriminant analysis model was established to show the correlation between spectral data and the measured SSC. Multiplicative scattering correction, standard normal variable, Savitzky-Golay convolution first derivative, and normalization (Normalize) were used for spectra preprocessing. Uninformative variables elimination, competitive adaptive reweighted sampling, and iteratively retaining informative variables were jointly used for wavelength optimization. NIR-based SSC prediction models for unpacked blueberries and PE-packaged blueberries were developed using partial least squares (PLS). The prediction model for PE-packaged samples (RP 2 =0.876, root mean square error of prediction [RMSEP]=0.632) had less precision than the model for unpacked samples (RP 2 =0.953, RMSEP=0.611). To reduce the effect of PE, the back propagation (BP) neural network and PLS were combined into the BP-PLS algorithm based on the residual learning algorithm. The model of BP-PLS (RP 2 =0.947, RMSEP=0.414) was successfully developed to improve the prediction accuracy of SSC for PE-packaged blueberries. The results suggested a promising way of using the BP-PLS method in tandem with NIR for the rapid detection of the SSC of PE-packaged blueberries. PRACTICAL APPLICATION: Most of the NIR-based research used unpacked blueberries as samples, while the use of packaged blueberries would provide researchers with a better understanding of the crucial factors at different phases of the blueberry supply chain (from harvest to sale). To meet market demands and minimize losses, NIR spectroscopy has been proven to be a rapid and nondestructive method for the determination of the SSC of PE-packaged blueberries. This study provides an effective method for monitoring the properties of blueberries in the entire supply chain.

The Prediction of Residual Electrical Life in Alternating Current Circuit Breakers Based on Savitzky-Golay-Long Short-Term.

In order to improve the accuracy of predicting the remaining electrical life of AC circuit breakers, ensure the safe operation of electrical equipment, and reduce economic losses caused by equipment failures, this paper studies a method based on the Savitzky-Golay convolution smoothing long short-term memory neural network for predicting the electrical life of AC circuit breakers. First, a full lifespan test is conducted to obtain degradation data throughout the entire life cycle of the AC circuit breaker, from which feature parameters that effectively reflect its operational state are extracted. Next, principal component analysis and the maximum information coefficient are used to remove redundancy in the feature parameters and choose the best subset of features. Subsequently, the Savitzky-Golay convolutional smoothing algorithm is employed to smooth the feature sequence, reducing the impact of noise and outliers on the feature sequence while preserving its main trends. Then, a secondary feature extraction is performed on the smoothed feature subset to obtain the optimal secondary feature subset. Finally, the remaining electrical lifespan of the AC circuit breaker is treated as a long-term sequence problem and the long short-term memory neural network method is used for precise time-series forecasting. The proposed model outperforms backpropagation neural networks and the gate recurrent unit in terms of prediction precision, achieving an impressive 97.4% accuracy. This demonstrates the feasibility of using time-series forecasting for predicting the residual electrical lifespan of electrical equipment and provides a reference for optimizing the method of predicting remaining electrical life.

Non-destructive determination of soluble solids content in intact apples using a self-made portable NIR diffuse reflectance instrument

The soluble solids content (SSC) is an important internal quality parameter of fruits that is monitored on the fruit market to meet different consumer demands. The aim of this study was to develop a self-made portable near-infrared (NIR) diffuse reflectance instrument to evaluate and monitor the SSC of intact apples. The visible and near-infrared (Vis-NIR) spectra of 118 ‘Fuji’ apples were collected using a Vis-NIR diffuse reflectance spectroscopic measurement system in the spectral range of 450–1100 nm. Savitzky-Golay convolution smoothing and first derivative methods were subsequently applied to eliminate noise interference and baseline shift. Successive projections algorithm (SPA) was performed to extract feature wavelengths from the pretreatment spectra, and the back-propagation artificial neural network (BP-ANN) and multivariate nonlinear regression (MNLR) models were employed to evaluate the predictive ability of the extracted and selected feature wavelengths for the SSC of apples. Seven wavelengths (881, 890, 901, 926, 941, 951, 978 nm) were selected as optimal feature wavelengths from the extracted feature wavelengths (18 wavelengths). The MNLR model (R2 = 0.953, RMSE = 0.391 %) exhibited a high prediction accuracy compared to the BP-ANN model (R2 = 0.865, RMSE = 0.754 %). Based on the results, NIR combined with the MNLR model could be applied in the self-made portable instrument to innovatively monitor the SSC of apples. Finally, a self-made portable NIR diffuse reflectance instrument was designed and developed, and the spectral information of 298 ‘Fuji’ apples was collected using the instrument. According to the results, the MNLR model could well predict the SSC of apples (R2 = 0.871, RMSE = 0.687 %). The overall results also revealed that the developed portable NIR instrument is a promising device for rapid non-destructive monitoring of fruit SSC, thereby meeting the practical application requirement of postharvest commercial processing systems. This instrument could also be beneficial to customers, growers, and producers.

Rapid determination of chemical compositions in chicken flesh by mining hyperspectral data

Hyperspectral data in 900–1700 nm range was mined for rapid and non-destructive determination of chicken chemical compositions. Hyperspectral images of chicken flesh samples were acquired and spectral data within the hyperspectral images were extracted, preprocessed to relate to moisture, protein and ash contents by partial least squares (PLS). The results showed that PLS model built with Savitzky-Golay convolution smoothing (SGCS) spectra (SGCS-PLS model), RAW spectra (RAW-PLS model), normalization (NOR) spectra (NOR-PLS model) had the best performance in predicting moisture (RP = 0.925, RMSEP = 0.391%), protein (RP = 0.926, RMSEP = 0.589%) and ash (RP = 0.938, RMSEP = 0.114%) contents, respectively. Four different methods including regression coefficients (RC), stepwise regression (SR), successive projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS) were then applied to select optimal wavelengths for simplifying the original PLS models. As a result, 23 and 15 optimal wavelengths were respectively selected by CARS method from SGCS spectra (for moisture) and NOR spectra (for ash) and the original PLS models were optimized (CARS-SGCS-PLS model, CARS-NOR-PLS model) with RP of 0.932, 0.932 and RMSEP of 0.376%, 0.120%, respectively. Thirteen optimal wavelengths were selected by RC method from RAW spectra (for protein) and RC-RAW-PLS model was constructed with RP of 0.923 and RMSEP of 0.588%. F test and t test results further verified the statistical soundness of the three best optimized models. The chemical maps of moisture, protein and ash were finally generated by transferring the models into each pixel of images of chicken samples to visualize the chemical distribution. The overall study indicated that hyperspectral data combined with PLS algorithm could be potentially used for predicting the chemical compositions of chicken flesh.

Rapid On-site identification of geographical origin and storage age of tangerine peel by Near-infrared spectroscopy

The feasibility of identifying geographical origin and storage age of tangerine peel was explored by using a handheld near-infrared (NIR) spectrometer combined with machine learning. A handheld NIR spectrometer (900–1700 nm) was used to scan the outer surface of tangerine peel and collect the corresponding NIR diffuse reflectance spectra. Principal component analysis (PCA) combined with Mahalanobis distance were used to detect outliers. The accuracies of all models in the anomaly set were much lower than that in calibration set and test set, indicating that the outliers were effectively identified. After removing the outliers, in order to initially explore the clustering characteristics of tangerine peels, PCA was performed on tangerine peels from different origins and the same origin with different storage ages. The results showed that the tangerine peels from the same origin or the same storage age had the potential to cluster, indicating that the spectral data of the same origin or the same storage age had a certain similarity, which laid the foundation for subsequent modeling and identification. However, there were quite a few samples with different origins or different storage ages overlapped and could not be distinguished from each other. In order to achieve qualitative identification of origin and storage age, Savitzky-Golay convolution smoothing with first derivative (SGFD) and standard normal variate (SNV) were used to preprocess the raw spectra. Random forest (RF), K-nearest neighbor (KNN) and linear discriminant analysis (LDA) were used to establish the discriminant model. The results showed that SGFD-LDA could accurately distinguish the origin and storage age of tangerine peel at the same time. The origin identification accuracy was 96.99%. The storage age identification accuracy was 100% for Guangdong tangerine peel and 97.15% for Sichuan tangerine peel. This indicated that the near-infrared spectroscopy (NIRS) combine with machine learning can simultaneously and rapidly identify the origin and storage age of tangerine peel on site.

Research on electrical life Trace prediction of contact based on SG-BP algorithm

In order to prevent electrical fire in low-voltage distribution system, it is a very effective measure to predict the electrical life of various low-voltage electrical appliances running on the line. At present, it is difficult to predict the residual electrical life of AC contactors. How to establish a reasonable and reliable prediction model for the residual electrical life of AC contactors is the key to the electrical life test and characteristic research of AC contactors. In this paper, Savitzky-Golay convolution smoothing algorithm is combined with BP neural network algorithm, and the remaining electrical life prediction model of contact is established by taking the arcing time, arcing energy and contact opening phase angle as the input parameters of the prediction model. The test results show that the average relative error of BP residual electrical life prediction model is less than 4%, and the maximum relative error is less than 10%. Combined with Savitzky-Golay convolution smoothing algorithm, SG-BP residual electrical life prediction model has average relative error less than 3% and maximum relative error less than 8%, which improves the prediction accuracy of residual electrical life, and lays a good foundation for making safe and reliable judgment by making better use of electrical life trace data..

Quick assessment of chicken spoilage based on hyperspectral NIR spectra combined with partial least squares regression

Pseudomonas spp. and Enterobacteriaceae are dominant spoilage bacteria in chicken during cold storage (0°C-4°C). In this study, high resolution spectra in the range of 900-1700 nm were acquired and preprocessed using Savitzky-Golay convolution smoothing (SGCS), standard normal variate (SNV) and multiplicative scatter correction (MSC), respectively, and then mined using partial least squares (PLS) algorithm to relate to the total counts of Pseudomonas spp. and Enterobacteriaceae (PEC) of fresh chicken breasts to predict PEC rapidly. The results showed that with full 900-1700 nm range wavelength, MSC-PLS model built with MSC spectra performed better than PLS models with other spectra (RAW-PLS, SGCS-PLS, SNV-PLS), with correlation coefficient (RP) of 0.954, root mean square error of prediction (RMSEP) of 0.396 log10 CFU/g and residual predictive deviation (RPD) of 3.33 in prediction set. Based on the 12 optimal wavelengths (902.2 nm, 905.5 nm, 923.6 nm, 938.4 nm, 946.7 nm, 1025.7 nm, 1124.4 nm, 1211.6 nm, 1269.2 nm, 1653.7 nm, 1691.8 nm and 1693.4 nm) selected from MSC spectra by successive projections algorithm (SPA), SPA-MSC-PLS model had RP of 0.954, RMSEP of 0.397 log10 CFU/g and RPD of 3.32, similar to MSC-PLS model. The overall study indicated that NIR spectra combined with PLS algorithm could be used to detect the PEC of chicken flesh in a rapid and non-destructive way. Keywords: hyperspectral NIR spectra, chicken, dominant spoilage, partial least squares regression, quick assessment DOI: 10.25165/j.ijabe.20211401.5726 Citation: Jiang S Q, He H J, Ma H J, Chen F S, Xu B C, Liu H, et al. Quick assessment of chicken spoilage based on hyperspectral NIR spectra combined with partial least squares regression. Int J Agric & Biol Eng, 2021; 14(1): 243–250.

Intelligent assessment of tea quality employing visible-near infrared spectra combined with a hybrid variable selection strategy

Tea is one of the most consumed and popular beverages around the world. Authenticating tea quality is essential to resist counterfeit goods and prevent fraud. This study innovatively aims to use a visible-near infrared (Vis-NIR) sensor combined with a variable combination population analysis (VCPA)-based hybrid variable selection algorithm to realize the authentication of tea quality. First, the spectra of 700 tea samples from seven different grades are acquired using a Vis-NIR sensor. Second, the raw spectra acquired are preprocessed by Savitzky-Golay convolution smoothing combined with first derivative (SG1), standard normal variate transformation (SNV), and multiplicative scatter correction (MSC), and the best pretreating technique is determined by support vector machine (SVM) algorithm. Third, the VCPA-based hybrid method optimizes the variable space of the optimal pretreated spectra data in the first step, and a genetic algorithm (GA) and iteratively retaining informative variables (IRIV) are utilized to execute further optimization and shrink the variable in the second step. Finally, authentication models of tea quality are constructed employing least squares support vector machine (LSSVM), random forest (RF), and back propagation neural network (BPNN) approaches based on the selected variables. The results show that the VCPA-IRIV-LSSVM model obtains the best predictive performance with an accuracy of 97.40% in the predictive process. Our findings demonstrate that it is feasible to apply the inexpensive vis-NIR sensor to assess quality attributes in tea. Also, the VCPA-based hybrid method can search the combination of the optimal wavelength and improve the identification accuracy and robustness of the model and can assist sensor manufacturers in developing a cheap and user-friendly smartphone-coupled micro spectroscopic sensor for authentication of tea quality.

The model updating based on near infrared spectroscopy for the sex identification of silkworm pupae from different varieties by a semi-supervised learning with pre-labeling method

It is effective to accurately discriminate the sex of silkworm pupae with the same varieties based on near infrared spectroscopy. However, when the model is promoted to classify new varieties of silkworm pupae, the model’s performance becomes worse, due to the cultivation environment and varieties changing. In the aims of improving the generalization ability and accuracy of the model, this paper proposed a model updating strategy based on semi-supervised learning. First, support vector machine identification model was built after the original spectra was pretreated by Savitzky-Golay convolution smoothing operation, which could effectively reduce spectra noise. Then, the support vector machine model gave the pre-labelings of unlabeled silkworm pupae in the updated set, which were divided into male samples and female samples. According to the correlation coefficients that calculated by Pearson correlation coefficient and Euclidean distance, a total of 8 reliable samples were selected from the male and female samples, respectively. The reliable samples were added to the original training set to update the original model. Finally, the updated model was used to test the test sets from the varieties of silkworm pupae that were the same with updated sets.The results showed the performance of the non-updated model for silkworm pupae from the three new varieties just reached 54.55%, 68.52%, 86.84%, respectively. The support vector machine model updated by using Pearson correlation coefficient improved the accuracy to 100%, 96.30%, 97.37%, and the model updated by Euclidean distance increased the identification accuracy of the three varieties that were not involved in the modeling to 100%, 75.93%, 92.10% respectively. The results showed that the performance of the model updated by Pearson correlation coefficient was better than Euclidean distance. The results revealed that the method based on semi-supervised learning could effectively solve the problem of poor universality for sex identification model.

A Novel Two-Step Spectral Recovery Framework for Coal Quality Assessment by Near-Infrared Spectroscopy

Near-infrared spectroscopy (NIRS) is an effective and efficient technique for evaluating coal quality. The original spectra might be contaminated by scattering interference and random noise. We propose a novel artifact removal framework to recover the buried information and to overcome limitations of currently available preprocessing techniques, such as the multiplicative scatter correction (MSC), as well as a smoothing process. The two-step framework is mainly constructed by MSC and Savitzky–Golay convolution (S-SGC) . Moreover, a particle swarm optimization (PSO) algorithm is used to search the optimal parameters within the framework. In addition, the spectra are collected from coal samples with different particle sizes (i.e., 0.2 and 3 mm), which may carry different characteristics and interfering information. We have analyzed seven kinds of coal properties, such as moisture (%), ash (%), volatile matter (%), and heating value (MJ/kg) via partial least square regression (PLSR) models in order to verify the effectiveness of the proposed method. The results show that the proposed two-step method provides superior performances for zooming in the spectral characteristic peaks and filtering the random noise simultaneously, which mainly benefits from the appropriate combination of MSC and S-SGC.

LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh

Total viable count (TVC) is often used as an important indicator for chicken freshness evaluation. In this study, 112 fresh chicken flesh samples were acquired after slaughtered and their hyperspectral images were collected in the LW-NIR (900-1700 nm) range. The full LW-NIR spectra (486 wavebands) within the images were extracted and applied to related to reference TVC values measured in different storage period, using partial least squares regression (PLSR) algorithm, resulting in high correlation coefficients (R) and low root mean square errors (RMSE), for either raw spectra or pretreatment spectra. By using regression coefficients (RC) method, 20, 18, 17 and 20 optimal wavebands were respectively selected from raw spectra, baseline correction (BC) spectra, Savitzky-Golay convolution smoothing (SGCS) spectra and standard normal variate (SNV) spectra and applied for the optimization of original full waveband PLSR model. By comparison, RC-PLSR model based on the SGCS spectra showed a better performance in TVC prediction with RC of 0.98 and RMSEC of 0.35 log10 CFU/g in calibration set, and RP of 0.98 and RMSEP of 0.44 log10 CFU/g in prediction set. At last, by transferring the best RC-PLSR model, the dynamic TVC change during the storage was visualized by color maps to indicate the TVC spoilage degree. The overall study revealed that LW-NIR hyperspectral imaging combined with PLSR could be used to predict the freshness of chicken flesh. Keywords: hyperspectral imaging, chicken, TVC, partial least square regression (PLSR) DOI: 10.25165/j.ijabe.20191203.4444 Citation: H Wang, H J He, H J Ma, F S Chen, Z L Kang, M M Zhu, et al. LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh. Int J Agric & Biol Eng, 2019; 12(3): 180–186.

LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh

Total viable count (TVC) is often used as an important indicator for chicken freshness evaluation. In this study, 112 fresh chicken flesh samples were acquired after slaughtered and their hyperspectral images were collected in the LW-NIR (900-1700 nm) range. The full LW-NIR spectra (486 wavebands) within the images were extracted and applied to related to reference TVC values measured in different storage period, using partial least squares regression (PLSR) algorithm, resulting in high correlation coefficients (R) and low root mean square errors (RMSE), for either raw spectra or pretreatment spectra. By using regression coefficients (RC) method, 20, 18, 17 and 20 optimal wavebands were respectively selected from raw spectra, baseline correction (BC) spectra, Savitzky-Golay convolution smoothing (SGCS) spectra and standard normal variate (SNV) spectra and applied for the optimization of original full waveband PLSR model. By comparison, RC-PLSR model based on the SGCS spectra showed a better performance in TVC prediction with RC of 0.98 and RMSEC of 0.35 log10 CFU/g in calibration set, and RP of 0.98 and RMSEP of 0.44 log10 CFU/g in prediction set. At last, by transferring the best RC-PLSR model, the dynamic TVC change during the storage was visualized by color maps to indicate the TVC spoilage degree. The overall study revealed that LW-NIR hyperspectral imaging combined with PLSR could be used to predict the freshness of chicken flesh. Keywords: hyperspectral imaging, chicken, TVC, partial least square regression (PLSR) DOI: 10.25165/j.ijabe.20191203.4444 Citation: H Wang, H J He, H J Ma, F S Chen, Z L Kang, M M Zhu, et al. LW-NIR hyperspectral imaging for rapid prediction of TVC in chicken flesh. Int J Agric & Biol Eng, 2019; 12(3): 180–186.

Nondestructive determination of the modulus of elasticity of Fraxinus mandschurica using near-infrared spectroscopy

A nondestructive methodology is proposed to determine the modulus of elasticity (MOE) of Fraxinus mandschurica samples by using near-infrared (NIR) spectroscopy. The test data consisted of 150 NIR absorption spectra of the wood samples obtained using an NIR spectrometer, with the wavelength range of 900 to 1900 nm. To eliminate the high-frequency noise and the systematic variations on the baseline, Savitzky–Golay convolution combined with standard normal variate and detrending transformation was applied as data pretreated methods. The uninformative variable elimination (UVE), improved by the evolutionary Monte Carlo (EMC) algorithm and successive projections algorithm (SPA) selected three characteristic variables from full 117 variables. The predictive ability of the models was evaluated concerning the root-mean-square error of prediction (RMSEP) and coefficient of determination (Rp2) in the prediction set. In comparison with the predicted results of all the models established in the experiments, UVE-EMC-SPA-LS-SVM presented the best results with the smallest RMSEP of 0.652 and the highest Rp2 of 0.887. Thus, it is feasible to determine the MOE of F. mandschurica using NIR spectroscopy accurately.

Substance identification based on transmission THz spectra using library search

Over recent years terahertz spectroscopy has become a new tool for the characterization of solid materials, in particular for investigating polymorphism and crystallinity in pharmaceutics. Search strategies have been tested for THz spectra of various organic compounds with their spectra taken from the Riken database (http://www.riken.jp), using the GRAMS spectroscopy software. A subset of the entire database was used, which had been processed by removal of atmospheric water vapour lines and smoothing applied based on Savitzky-Golay convolution or time domain filtering. The spectral range available for all library substances was restricted to an interval from 0.9 to 4.5 THz (30–150cm−1). The number of vibrational bands within this spectral range is much reduced compared with mid-infrared or Raman spectra. The appropriateness of spectral pre-treatment is demonstrated with regard to reliability and robustness of the search methods. In particular, time-domain filters for smoothing and pre-treatment by the removal of water lines and etalon effects have been successfully tested in combination with least squares and correlation methods. With these insights, applications for substance identification, especially for the pharmaceutical industry, may be enlarged.

Numerical differentiation methods for the logarithmic derivative technique used in dielectric spectroscopy

In dielectric relaxation spectroscopy the conduction contribution often hampers the evaluation of dielectric spectra, especially in the low-frequency regime. In order to overcome this the logarithmic derivative technique could be used, where the calculation of the logarithmic derivative of the real part of the complex permittivity function is needed. Since broadband dielectric measurement provides discrete permittivity function, numerical differentiation has to be used. Applicability of the Savitzky-Golay convolution method in the derivative analysis is examined, and a detailed investigation of the influential parameters (frequency, spectrum resolution, peak shape) is presented on synthetic dielectric data.

Simultaneous determination of catalysts through a single catalytic kinetic run

A method for the simultaneous determination of catalysts through a single kinetic run is proposed. It is based on the difference in the apparent orders pi of the indicator reaction catalysed by each catalyst. The concentration of each catalyst can be deduced by multiple linear regression of the reaction rate v versus Api, where A is the concentration of the reactant, which changes with time. The Savitzky—Golay convolution method is used for calculating the reaction rate. As an example, Ru and Os have been determined on the basis of their catalytic effects on the indicator reaction between CeIV and AsIII.