==== Front Food Chem X Food Chem X Food Chemistry: X 2590-1575 Elsevier S2590-1575(23)00109-8 10.1016/j.fochx.2023.100667 100667 Article(s) from the Special Issue on Food Authentication and Origin by Dr. Yong Fang and Dr. Vural Gökmen Comparison of a portable Vis-NIR hyperspectral imaging and a snapscan SWIR hyperspectral imaging for evaluation of meat authenticity Dashti Abolfazl abc Müller-Maatsch Judith a Roetgerink Emma a Wijtten Michiel a Weesepoel Yannick yannick.weesepoel@wur.nl a⁎ Parastar Hadi d Yazdanpanah Hassan ce a Wageningen Food Safety Research, Wageningen University and Research, Wageningen, the Netherlands b Forensic Toxicology Department, Legal Medicine Research Center, Legal Medicine Organization, Tehran, Iran c Department of Toxicology and Pharmacology, School of Pharmacy, Shahid Beheshti University of Medical Sciences, Tehran, Iran d Department of Chemistry, Sharif University of Technology, Tehran, Iran e Food Safety Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran ⁎ Corresponding author. yannick.weesepoel@wur.nl 03 4 2023 30 6 2023 03 4 2023 18 10066731 10 2022 10 3 2023 28 3 2023 © 2023 The Authors 2023 https://creativecommons.org/licenses/by/4.0/ This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Highlights • Portable Vis-NIR and SWIR hyperspectral imaging were used for meat authenticity. • Linear and non-linear chemometrics methods were used to analyze the data. • The best results were achieved with Vis-NIR HSI combined with non-linear methods. • Vis-NIR and SWIR HSI may be used for online/in-line routine screening of meat authenticity. The performance of visible-near infrared hyperspectral imaging (Vis-NIR-HSI) (400–1000 nm) and shortwave infrared hyperspectral imaging (SWIR-HSI) (1116–1670 nm) combined with different classification and regression (linear and non-linear) multivariate methods were assessed for meat authentication. In Vis-NIR-HSI, total accuracies in the prediction set for SVM and ANN-BPN (the best classification models) were 96 and 94 % surpassing the performance of SWIR-HSI with 88 and 89 % accuracy, respectively. In Vis-NIR-HSI, the best-obtained coefficient of determinations for the prediction set (R2p) were 0.99, 0.88, and 0.99 with root mean square error in prediction (RMSEP) of 9, 24 and 4 (%w/w) for pork in beef, pork in lamb and pork in chicken, respectively. In SWIR-HSI, the best-obtained R2p were 0.86, 0.77, and 0.89 with RMSEP of 16, 23 and 15 (%w/w) for pork in beef, pork in lamb and pork in chicken, respectively. The results ascertain that Vis-NIR-HSI coupled with multivariate data analysis has better performance rather than SWIR-HIS. Keywords Portable HSI Snapscan HSI Meat authenticity PCA Chemometrics ==== Body pmc1 Introduction Meat authenticity and meat fraud detection are challenging for scientists, industries, regulatory agencies, and consumers. Meat authenticity is, however, important regarding different social and health issues, in particular for consumers following a religious diet (Halal, kosher, etc.) and who suffer from allergies (Dashti et al., 2021; Mohammed Kamruzzaman, 2021). Also, the authenticity of food products affects the credibility of the companies. Different analytical methods including, polymerase chain reaction (PCR) (Kumar, 2021), chromatographic and mass spectrometry (MS) techniques (Stachniuk et al., 2021), and immunoassay-based methods (Mandli et al., 2018) were developed for meat authentication. Current approaches for meat fraud detection have different limitations such as being time-consuming, laborious, expensive, and destructive. In addition, they usually require high-skill technicians (Dashti et al., 2021; Mohammed Kamruzzaman, 2021; Mohammed Kamruzzaman et al., 2016). Hence, there is a need for developing new non-invasive, rapid and accurate detection techniques, especially to design online/portable detection instruments/devices that can be applied throughout the food chain (Zhao et al., 2019). Hyperspectral imaging integrates both spectroscopic and imaging techniques in one system. The combined nature of techniques provides simultaneously physical and spatial characteristics such as shape, size, appearance and color of the sample under analysis and intrinsic chemical and molecular information thereof through spectral analysis (Wu and Sun, 2013a, Wu and Sun, 2013b). Among the available technologies, hyperspectral imaging (HSI) is a relatively new technology that has provided a promising alternative, offering speed, accuracy, and reliability, besides being a nondestructive analysis technology for different production processes (Siche et al., 2016). Recently, HSI techniques have received much attention for application in non-destructive food analysis and therefore also in meat and meat products. These techniques have been successfully implemented for meat quality (Jia et al., 2022; Mohammed Kamruzzaman et al., 2013, Zhang et al., 2019), offal adulteration detection in ground meat (Grundy et al., 2022, Jiang et al., 2021; M Kamruzzaman et al., 2014) and meat speciation (Al-Sarayreh et al., 2017, Jiang et al., 2019; Mohammed Kamruzzaman et al., 2012; Mohammed Kamruzzaman et al., 2013, Rady and Adedeji, 2020). Previous studies have mainly used benchtop hyperspectral instruments for the application of meat authenticity. Considering the large amount of adulteration of meat and the need for rapid detection at various stages of the meat supply chain, there is an urgent need to develop non-destructive, fast and portable techniques for on/in site analysis of meat authenticity at various stages of meat supply chain. The objective of this study was: (a) to compare the practicability of two hyperspectral imaging systems with different wavelength ranges including portable visible-near infrared (400–1000 nm) and a snapscan short wave infrared (SWIR, 1116–1670 nm) in combination with different linear and non-linear classification multivariate data analysis to identify authentication of different meat species (lamb, beef, chicken and pork); (b) to quantify the pork in lamb, beef and chicken samples with dereferent linear and non-linear regression methods. Finally,this research can provide best wavelength (regarding with these cameras) and multivariate methods for meat speciation behind inherent features of portable and snapscan HSI. 2 Material and methods 2.1 Collection and preparation of meat samples Lamb (n = 40), beef (n = 40), chicken (n = 40), and pork (n = 40) meat samples (in total, 160 samples) were purchased from local butchers and supermarkets in the Netherlands over four months in 2020 (Table S1). Species-specific PCR was used as a reference method to verify the species of the samples (Scholtens et al., 2017). This method was selected due to its high sensitivity and relative rapid analyses. Samples that tested negative were removed from the dataset. Mixtures of meat samples (approximately 300 g) were prepared by manually removing skin and visible fat with a kitchen knife, cut into small cubes (2x2x2 cm), and ground separately using a meat grinder (Tristar VM-4210, Smartwares group, Tilburg, The Netherlands) with the coarse grinding disc (hole diameter 0.7 cm). After grinding, parts of the ground meat samples were mixed by hand. The mixtures were prepared by adding 2 %, 5 %, 10 %, 25 %, and 50 % pork (w/w) to lamb, beef, and chicken samples. Homogeneity was checked visually. Mixtures were prepared by randomization of samples (Table S2). Samples were vacuum-sealed using and stored at −18.5 °C in the dark and conditioned at room temperature for 12 h in sealed containers before data acquisition. 2.2 Data acquisition 2.2.1 Vis-NIR (400–1000 nm) spectral imaging Data acquisition in the Vis-NIR was performed using a portable Specim IQ hyperspectral line-scan camera with push-broom technology (Specim, Spectral Imaging ltd., Oulu, Finland). Image acquisition was performed in the wavelength range of 400–1000 nm in reflectance mode. The spatial resolution of the recorded data was 512x512 pixels and the spectral resolution was 7 nm resulting in 204 spectral bands across the wavelength range. A halogen-based illumination (2 × 1250 W) that covers the full wavelength range of 400 to 1000 nm range was used. A 95 % white reference tile was used for calibration before data acquisition. 2.2.2 SWIR (1116–1670 nm) spectral imaging An IMEC SWIR hyperspectral Snapscan imaging camera (Interuniversity Microelectronics Centre, Leuven, Belgium) with 108 spectral bands and a spectral range of 1116–1670 nm equipped with an Optec 16 mm F1.7 SWIR lens (Optec S.p.A., Parabiago, Italy) was used. The system was controlled by the IMEC Snapscan software (version 1.3.0.8, IMEC). Four small halogen lamps were utilized for the homogeneous illumination of the samples. Before data acquisition, the camera was calibrated using a 95 % reflection calibration white standard (WS) tile sized 200 × 200 mm, using an integration time of 2.5 ms for the 640 × 512 × 108 (x * y * λ) hypercubes. 2.3 Data analysis Background segmentation, extraction of spectral data from the hyperspectral images and calculations performed in the present research were carried out using MATLAB (R2017b, The Mathworks Inc.), PLS ToolBox (R9.0, Eigenvector Inc.), and MIA_Toolbox (version 3.1, Eigenvector Inc.). ROIs from the Vis-NIR data were selected by applying the mask. The mask was created by subtracting a low-reflectance band from a high-reflectance band. ROIs from the SWIR images were selected by manual removal of the background using the MIA toolbox. The mean spectra of ROIs were calculated by averaging the spectra of all pixels within the ROI at each wavelength. 2.3.1 Exploring the pattern in pure meat samples Principal component analysis (PCA) was used to indicate clustering patterns of the samples (PCA score plots) and to assess wavelengths that have the most influence on the classification of samples (PCA loading plots) (Abdi & Williams, 2010). 2.3.2 Discrimination (classification) of pure meat samples For species classification of samples linear and non-linear chemometric models were tested: partial least-squares discriminant analysis (PLS-DA) (Chevallier et al., 2006), support vector machine (SVM) (Jakkula, 2006)with radial basis function and artificial neural networks based on backpropagation network (ANN-BPN) (Wythoff, 1993). Venetian blinds cross-validation was used to assess the performance of the model and minimize the risk of overfitting data. In PLS-DA, the optimal number of PLS factors (latent variables (LVs)) for the models was selected by Venetian blinds cross-validation (number of data split: 10, thickness: 1) according to minimum RMSEC and RMSECV plots. The performance of classification models was evaluated according to the percentage of samples truly classified during calibration development and, afterward, with external validation. These performance parameters, such as sensitivity, specificity, accuracy and error rate are usually derived from a confusion matrix, to better assess the classification performance. These parameters were calculated according to the subsequent expressions combining the number of true positives (TP, correctly identified), true negatives (TN, correctly rejected), false positives (FP, incorrectly identified) and false negatives (FN, incorrectly rejected) achieved in calibration and validation. These parameters are defined according to Equations (1), (2), (3), (4) (Ballabio and Consonni, 2013, Tharwat, 2018).(1) Sensitivity=TPTP+FN (2) Specificity=TNTN+FP (3) Accuracy=TP+TNTP+TN+FP+FN (4) Errorrate=1-Acc=FP+FNTP+TN+FP+FN 2.3.3 Regression methods For identification of the presence of port in meat samples, different regression linear and non-linear chemometric models were tested: partial least squares (PLSR), support vector machines regressing (SVM-R) and artificial neural networks based on multi-layer back-propagation perceptron (ANN-MLP) were applied to predict the percentage of adulteration. 2.4 Software Clustering (PCA), classification (PLS-DA, SVM and ANN-BPN) and regression (PLSR and SVM-R) analysis were performed using PLS-toolbox version 9.0 (Eigenvector, USA). Also, ANN-MLP regression model was applied using MVC1 toolbox (Olivieri et al., 2004) written for MATLAB. Before chemometric modeling, an appropriate data pretreatment was chosen based on the performance of the chemometric model. Effects of the different data pretreatments are displayed in the results and discussion section. A flowchart that explains the complete analysis of the hyperspectral data starting from image acquisition to multivariate analysis is shown in Figure S1. 3 Results and discussion Fig. 1 shows Vis-NIR and SWIR averaged spectra obtained from four pure meat species. According to the heavy overlap of the spectra for different samples, chemometric analysis is necessary for the extraction of useful information.Fig. 1 A) Vis-NIR and B) SWIR averaged spectra obtained from all four pure meat species. 3.1 Exploring the pattern in pure meat samples Different preprocessing techniques were tested and the Gap segment 2nd order derivative with external parameter orthogonalization (EPO) (2 PCs) and mean center data preprocessing steps were applied before PCA. These pre-processing methods were found preferable for both Vis-NIR and SWIR HSI cameras, respectively (data of other pre-processing tested not displayed). According to the PCA score plots in Fig. 2 A and B, separations between groups in score space for both cameras were found when classifying the four different meat species. In the Vis-NIR PCA score plot, PC1 is corresponding to the maximum variance capture and provides the best separations (clustering) between species. The highest loading on PC1 and PC2 (Fig. 2 C) were found around 418, 550 and 578 nm associated with meat pigments and wavelength between 800 and 1000 nm could be related to meat composition (water, protein and fat,..). On contrary, PC1 is corresponding to the maximum variance but the best clustering was on PC2 in the SWIR PCA score plot. According to the loading on PC1 and PC2 (Fig. 2 D), the wavelengths between 1116 and 1500 nm (associated with the water and protein content) have more contributions to samples clustering (Cozzolino and Murray, 2004, Downey et al., 2000).Fig. 2 The PCA score plots of Vis-NIR (A) and SWIR (B) and PCA loading plot of Vis-NIR spectra preprocessed with Gap segment 2nd order derivative with external parameter orthogonalization (EPO) (2 PCs) (C) and SWIR spectra preprocessed with mean center (D) for pure meat samples. The better separation between the species in PCA score plots of Vis-NIR spectra compared with SWIR spectra may be caused by the different wavelength ranges. In the Vis-NIR features, the myoglobin and hemoglobin content of the different meat species can be incorporated into the information, while in the SWIR more information on the macro-composition like moisture, protein, and fat is detected. Therefore, the adequate separation between species in Vis-NIR HSI rather than SWIR HSI could be related to both pigment color (visible region) and muscle composition (NIR region) information. These results are in agreement with those reported by other authors (Alomar et al., 2003, Cozzolino and Murray, 2004; Mohammed Kamruzzaman et al., 2016, Moran et al., 2018). Interestingly, the type of cut was not found to be relevant in the PCA analysis. Furthermore, the wavelengths with a higher value are important to explain the variance and are potential wavelengths to differentiate the four species (see Fig. 2 C and D). These wavelengths can be used for meat classification instead of the whole spectral range. 3.2 Discrimination (classification) of pure meat samples The most appropriate pre-processing strategy was chosen according to the highest sensitivity, specificity, and lowest classification error. The best performing PLS-DA models for Vis-NIR and SWIR cameras were achieved with baseline correction in addition to 1st derivative (order: 2, window 19pt) and 2nd derivative (order: 2, window 21pt), and OSC (orthogonal signal correction), respectively (Table 1, Table 2).Table 1 Classification performance (in %) of PLS-DA, SVM (kernel function: RBF), and ANN (BPN algorithm) models for classification of lamb, beef, chicken, and pork with Vis-NIR hyperspectral camera (400–1000 nm). PLS-DA1 SVM2 ANN3 Train CV Test Train CV Test Train CV Test Lamb Sensitivity 95.5 91.0 87.5 100.0 95.5 87.5 100.0 86.4 87.5 Specificity 88.5 85.2 82.1 100.0 96.7 96.4 100.0 98.4 92.9 Accuracy 92.0 88.1 84.9 100.0 96.1 92.0 100.0 92.4 90.2 Error 8.0 11.9 15.1 0.0 3.9 8.0 0.0 7.6 9.8 Beef Sensitivity 100.0 100.0 77.8 100.0 100.0 88.9 100.0 100.0 88.9 Specificity 98.4 92.1 88.9 100.0 100.0 96.3 100.0 98.4 96.3 Accuracy 99.2 96.1 83.4 100.0 100.0 92.6 100.0 99.3 92.6 Error 0.8 3.9 16.6 0.0 0.0 7.4 0.0 0.7 7.4 Chicken Sensitivity 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Specificity 100.0 96.7 100.0 100.0 98.3 100.0 100.0 100.0 100.0 Accuracy 100.0 98.4 100.0 100.0 99.2 100.0 100.0 100.0 100.0 Error 0.0 1.6 0.0 0.0 0.8 0.0 0.0 0.0 0.0 Pork Sensitivity 83.3 72.2 91.7 100.0 83.3 100.0 100.0 61.1 91.7 Specificity 83.3 80.0 91.7 100.0 98.5 100.0 100.0 87.7 100.0 Accuracy 83.2 76.2 91.7 100.0 90.9 100.0 100.0 74.5 95.9 Error 16.8 23.8 8.3 0.0 9.1 0.0 0.0 25.5 4.1 1 preprocessed with baseline correction + 1st derivative (order: 2, window 19pt). 2 preprocessed with 1st derivative (order: 2, window: 15pt). 3 preprocessed with SNV. Table 2 Classification performance (in %) of PLS-DA, SVM (kernel function: RBF), and ANN (BPN algorithm) models for classification of lamb, beef, chicken, and pork with SWIR hyperspectral camera (1116–1670 nm). PLS-DA1 SVM2 ANN3 Train CV Test Train CV Test Train CV Test Lamb Sensitivity 92.6 88.9 83.3 77.8 70.4 58.3 88.9 77.8 83.3 Specificity 65.5 58.3 47.2 97.6 90.5 91.7 91.7 78.6 86.1 Accuracy 79.1 73.7 65.3 87.7 80.4 75.0 90.3 78.2 84.8 Error 20.9 26.3 34.7 12.3 19.6 25.0 9.7 21.8 15.2 Beef Sensitivity 100.0 95.5 100.0 95.5 86.4 94.1 95.5 90.9 88.2 Specificity 94.4 88.8 100.0 100.0 100.0 96.8 96.6 93.3 93.5 Accuracy 97.2 92.2 100.0 97.7 93.2 95.4 96.1 92.1 90.9 Error 2.8 7.8 0.0 2.3 6.8 4.6 3.9 7.9 9.1 Chicken Sensitivity 93.1 86.2 100.0 89.7 86.2 100.0 100.0 89.7 100.0 Specificity 85.4 79.3 83.8 97.6 90.2 100.0 100.0 91.5 100.0 Accuracy 89.3 82.8 91.9 93.6 88.2 100.0 100.0 90.6 100.0 Error 10.7 17.2 8.1 6.4 11.8 0.0 0.0 9.4 0.0 Pork Sensitivity 93.9 90.9 100.0 87.9 60.6 75.0 97.0 75.8 75.0 Specificity 38.5 41.0 52.5 87.2 84.6 90.0 97.4 74.4 90.0 Accuracy 66.3 66.0 76.3 87.5 72.6 82.5 96.6 75.1 82.5 Error 33.7 34.0 23.7 12.5 27.4 17.5 3.4 24.9 17.5 1 preprocessed with 2nd derivative (order: 2, window 21pt) and OSC (orthogonal signal correction). 2 preprocessed with 2nd derivative (order: 2, window: 9pt) and Pareto (Sqrt Std) scaling. 3 preprocessed with SNV (standard normal variate). In this regard, the best total accuracy values for Vis-NIR and SWIR were 90 % and 83 %, respectively. Also in the SVM method, the best appropriate pre-processing strategy was chosen according to the highest sensitivity, specificity, and lowest classification error. Hereof, 1st derivative (order: 2, window: 15pt) and 2nd derivative (order: 2, window: 9pt) plus Pareto scaling(scale each variable by the square root of its standard deviation (Sqrt Std) were the most proper pre-processing methods for Vis-NIR and SWIR, respectively. The best total accuracy values for Vis-NIR and SWIR were 96 % and 88 %, receptively (Table 1, Table 2). For ANN-BPN methods, the best results were achieved with SNV for both cameras and the best total accuracy was 94 % and 89 % for Vis-NIR and SWIR, respectively (Table 1, Table 2). Table 1, Table 2 show that non-linear models like SVM with RBF kernel and ANN-BPN, have a better prediction ability than the linear models (PLS-DA) for both HSI systems. The total accuracy for Vis-NIR was arranged between 90 and 96 % and the total accuracy for SWIR was 83–89 %. In our work, the Vis-NIR camera has superior performance for pure meat classification of lamb, beef, chicken, and pork in comparison with the SWIR camera. It could be due to the different and longer wavelength range of the Vis-NIR hyperspectral camera (400–1000 nm) rather than the SWIR hyperspectral camera (1116–1670 nm). Visual differences were observed between sample species in both visible and NIR regions (Fig S1 A). Peaks around 418, 546, and 578 nm may be related to hemoglobin and myoglobin absorption. This showed that both pigment color information (from the visible region) and the composition of the meat (from the NIR region) gave information to be used for speciation purposes (Alamprese et al., 2013, Cozzolino and Murray, 2004, Dashti et al., 2021, Moran et al., 2018). Snapscan HSI systems have limited spectral information for red-meat speciation and authentication while proposing advantages such as portability and video rate imaging (Al-Sarayreh et al., 2020). 3.3 Detection of adulteration levels of pork in the meat of other species The main objective of this contribution was to quantify the level of adulteration (pork) in minced lamb, beef, and chicken. The regression models were developed using PLS, SVM, and ANN-MLP to correlate the spectral data with the level of adulteration. Here different spectral pre-treatments were assessed for all these models and the best models were selected according to the highest R2 and lowest RMSE for calibration and cross-validation. Full cross-validation and test set were utilized to validate internally and externally all predictive models. As illustrated in Table 3, overall, the performance of the ANN-MLP regression model (as a fully non-linear model) in the case of cross-validation and prediction was better than PLS and SVM regression models. In Vis-NIR hyperspectral camera, the best quantification models belong to the detection of pork in beef, pork in chicken, and the end, pork in lamb, respectively. The weaker results for the quantification of pork in lamb could be due to the similarity between their spectra (Fig. S2 A).Table 3 Performance comparison of the three regression models for quantification of pork in beef, lamb, and chicken meat samples (without preprocessing and with the best preprocessing) with Vis-NIR hyperspectral camera (400–1000 nm). Preprocessing LVs RMSEC RMSECV RMSEP R2C R2CV R2P PLS-R Pork in beef Non 5 12 15 17 0.83 0.74 0.83 2nd derivative 4 10 12 14 0.90 0.83 0.90 Pork in lamb Non 4 16 20 22 0.79 0.68 0.75 2nd derivative 5 14 18 24 0.84 0.73 0.76 Pork in chicken Non 4 14 17 20 0.86 0.80 0.57 Gap segment 2nd derivative 8 8 17 13 0.95 0.81 0.82 SVM-R Pork in beef Non ---- 13 17 21 0.82 0.68 0.87 EMSC ---- 9 13 9 0.91 0.81 0.95 Pork in lamb Non ---- 16 20 23 0.81 0.70 0.80 ---- ---- ---- ---- ---- ---- ---- ---- Pork in chicken Non ---- 12 20 17 0.90 0.74 0.73 Baseline correction ---- 7 15 10 0.97 0.83 0.86 ANN-MLP Pork in beef Non ---- ---- 6 15 ---- 0.98 0.98 Detrend ---- ---- 6 9 ---- 0.98 0.99 Pork in lamb Non ---- ----- 22 27 ---- 0.78 0.87 Centering ---- ----- 20 24 ---- 0.81 0.88 Pork in chicken Non ---- ----- 5 10 ---- 0.99 0.94 Detrend ---- ----- 5 4 ---- 0.99 0.99 Latent variables (LVs), Root Mean Squared Error of Calibration (RMSEC), Root Mean Squared Error of Cross-Validation (RMSECV), Root Mean Squared Error of Prediction (RMSEP). The results revealed that spectral pre-treatment methods offer more improvement in the performance of PLS and SVM models. In ANN models, spectral pre-treatment methods mainly affected RMSE rather than R square (Table 3). As can be seen in Table 4, the performance of the ANN-MLP regression model in the case of cross-validation and prediction was better than PLS and SVM regression models. Here, different spectral preprocessing methods were tried for SWIR spectra, but unfortunately, no suitable PLS and SVM regression models could be gained with those spectra. According to Table 3, Table 4, overall, the Vis-NIR camera had a better performance for pork meat adulteration quantification in other species (lamb, beef, and chicken).Table 4 Performance comparison of the three regression models for quantification of pork in beef, lamb, and chicken meat samples (without preprocessing and with the best preprocessing) with SWIR hyperspectral camera (1116–1670 nm). Preprocessing LVs RMSEC RMSECV RMSEP R2C R2CV R2P PLS-R Pork in beef Non 4 22 26 19 0.64 0.49 0.53 1st derivative + baseline correction 6 17 23 22 0.78 0.62 0.54 Pork in lamb Non 2 33 34 27 0.18 0.12 0.24 EMSC 4 27 32 29 0.43 0.21 0.19 Pork in chicken Non 3 27 31 29 0.30 0.12 0.48 Gap segment 2nd derivative 8 17 27 24 0.75 0.44 0.56 SVM-R Pork in beef Non ---- 20 24 15 0.73 0.58 0.69 SNV ---- 10 19 18 0.92 0.73 0.67 Pork in lamb Non ----- 24 27 22 0.59 0.44 0.54 Pareto (Sqrt Std) scaling ----- 17 23 21 0.79 0.61 0.61 Pork in chicken Non ----- 40 40 36 0.01 0.64 0.02 OSC ----- 37 37 34 0.55 0.35 0.27 ANN-MLP Pork in beef Non ---- ---- 15 16 ---- 0.91 0.86 -------- ---- ---- ---- ---- ---- ---- ---- Pork in lamb Non ----- ----- 24 25 ----- 0.75 0.65 MSC ----- ----- 26 23 0.72 0.77 Pork in chicken Non ----- ----- 25 27 ---- 0.75 0.55 2nd derivative (Sav Gol) ----- ----- 14 15 ---- 0.92 0.89 Latent variables (LVs), Root Mean Squared Error of Calibration (RMSEC), Root Mean Squared Error of Cross-Validation (RMSECV), Root Mean Squared Error of Prediction (RMSEP). 4 Conclusions This contribution was carried out to compare the potential of two HSI cameras for speciation and adulteration quantification of different meat species combined with linear and non-linear classifications (PLS-D, SVM, and ANN-BPN) and regressions (PLSR, SVM-R, and ANN-MLP) multivariate methods. In classification methods, the total accuracy for Vis-NIR was between 90 and 96 % and the total accuracy for SWIR was 83–89 %. In this approach, the best performances were acquired with ANN-BPN in both HSI cameras. For pork quantification in other species, the ANN-MLP regression method showed better performance in terms of cross-validation and prediction in both HSI systems rather than PLS and SVM regressions methods, and the best performance was acquired with the portable Vis-NIR HSI camera. The results demonstrated that the spectral data collected from portable Vis-NIR hyperspectral imaging combined with non-linear multivariate methods provide better performance for meat speciation and detection of pork adulteration in minced lamb, beef, and chicken rather than snapscan SWIR hyperspectral imaging. Funding This work was supported by the Dutch Ministry of Agriculture, Nature and Food Quality (Knowledge base grant, KB-38–001-008). Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Appendix A Supplementary data The following are the Supplementary data to this article:Supplementary data 1 Data availability The authors do not have permission to share data. Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2023.100667. ==== Refs References Abdi H. Williams L.J. Principal component analysis Wiley Interdisciplinary Reviews: Computational Statistics 2 4 2010 433 459 Al-Sarayreh M. Reis M.M. Yan W.Q. Klette R. Potential of deep learning and snapshot hyperspectral imaging for classification of species in meat Food Control 117 2020 107332 Al-Sarayreh M. Reis M.M. Yan W.Q. Klette R. Detection of adulteration in red meat species using hyperspectral imaging Pacific-Rim Symposium on Image and Video Technology 2017 182 196 Alamprese C. Casale M. Sinelli N. Lanteri S. Casiraghi E. Detection of minced beef adulteration with turkey meat by UV–vis, NIR and MIR spectroscopy LWT-Food Science and Technology 53 1 2013 225 232 Alomar D. Gallo C. Castaneda M. Fuchslocher R. Chemical and discriminant analysis of bovine meat by near infrared reflectance spectroscopy (NIRS) Meat Science 63 4 2003 441 450 22062513 Ballabio D. Consonni V. Classification tools in chemistry. Part 1: Linear models. PLS-DA Analytical Methods 5 16 2013 3790 3798 Chevallier S. Bertrand D. Kohler A. Courcoux P. Application of PLS-DA in multivariate image analysis Journal of Chemometrics: A Journal of the Chemometrics Society 20 5 2006 221 229 Cozzolino D. Murray I. Identification of animal meat muscles by visible and near infrared reflectance spectroscopy LWT-Food Science and Technology 37 4 2004 447 452 Dashti A. Müller-Maatsch J. Weesepoel Y. Parastar H. Kobarfard F. Daraei B. …Yazdanpanah H. The Feasibility of Two Handheld Spectrometers for Meat Speciation Combined with Chemometric Methods and Its Application for Halal Certification Foods 11 1 2021 71 https://www.mdpi.com/2304-8158/11/1/71 35010197 Downey G. McElhinney J. Fearn T. Species identification in selected raw homogenized meats by reflectance spectroscopy in the mid-infrared, near-infrared, and visible ranges Applied Spectroscopy 54 6 2000 894 899 Grundy H.H. Brown L. Romero M.R. Donarski J. Methods to determine offal adulteration in meat products to support enforcement and food security Food Chemistry 133818 2022 Jakkula, V. (2006). Tutorial on support vector machine (svm). School of EECS, Washington State University, 37(2.5), 3. Jia W. van Ruth S. Scollan N. Koidis A. Hyperspectral imaging (HSI) for meat quality evaluation across the supply chain: Current and future trends Current Research in Food Science. 2022 Jiang H. Ru Y. Chen Q. Wang J. Xu L. Near-infrared hyperspectral imaging for detection and visualization of offal adulteration in ground pork Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 249 2021 119307 Jiang H. Wang W. Zhuang H. Yoon S.-C. Yang Y. Zhao X. Hyperspectral imaging for a rapid detection and visualization of duck meat adulteration in beef Food Analytical Methods 12 2019 2205 2215 Kamruzzaman M. Haque M.E. Ali M.R. Hyperspectral imaging technique for offal quantification in minced meat Journal of the Bangladesh Agricultural University 12 1 2014 189 194 Kamruzzaman M. Fraud Detection in Meat Using Hyperspectral Imaging Meat and Muscle Biology 5 3 2021 Kamruzzaman M. Barbin D. ElMasry G. Sun D.-W. Allen P. Potential of hyperspectral imaging and pattern recognition for categorization and authentication of red meat Innovative Food Science & Emerging Technologies 16 2012 316 325 Kamruzzaman M. ElMasry G. Sun D.-W. Allen P. Non-destructive assessment of instrumental and sensory tenderness of lamb meat using NIR hyperspectral imaging Food Chemistry 141 1 2013 389 396 23768372 Kamruzzaman M. Makino Y. Oshita S. Rapid and non-destructive detection of chicken adulteration in minced beef using visible near-infrared hyperspectral imaging and machine learning Journal of Food Engineering 170 2016 8 15 Kamruzzaman M. Sun D.-W. ElMasry G. Allen P. Fast detection and visualization of minced lamb meat adulteration using NIR hyperspectral imaging and multivariate image analysis Talanta 103 2013 130 136 23200368 Kumar Y. Isothermal amplification-based methods for assessment of microbiological safety and authenticity of meat and meat products Food Control 121 2021 107679 Mandli J. Fatimi I.E.L. Seddaoui N. Amine A. Enzyme immunoassay (ELISA/immunosensor) for a sensitive detection of pork adulteration in meat Food Chemistry 255 2018 380 389 29571490 Moran L. Andres S. Allen P. Moloney A.P. Visible and near infrared spectroscopy as an authentication tool: Preliminary investigation of the prediction of the ageing time of beef steaks Meat Science 142 2018 52 58 29660544 Olivieri A.C. Goicoechea H.C. Iñón F.A. MVC1: An integrated MatLab toolbox for first-order multivariate calibration Chemometrics and Intelligent Laboratory Systems 73 2 2004 189 197 Rady A. Adedeji A.A. Application of hyperspectral imaging and machine learning methods to detect and quantify adulterants in minced meats Food Analytical Methods 13 4 2020 970 981 Scholtens I.M.J. Prins T.W. Van Raamsdonk L.W.D. Specificity of a novel TaqMan PCR method for detection of poultry DNA Food Control 73 2017 532 539 Siche R. Vejarano R. Aredo V. Velasquez L. Saldana E. Quevedo R. Evaluation of food quality and safety with hyperspectral imaging (HSI) Food Engineering Reviews 8 3 2016 306 322 Stachniuk A. Sumara A. Montowska M. Fornal E. Peptide markers for distinguishing guinea fowl meat from that of other species using liquid chromatography–mass spectrometry Food Chemistry 345 2021 128810 Tharwat, A. (2018). Classification assessment methods. Applied Computing and Informatics (2018). URL: Http://Www. Sciencedirect. Com/Science/Article/Pii S. Wu D. Sun D.-W. Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review—Part I: Fundamentals Innovative Food Science & Emerging Technologies 19 2013 1 14 Wu D. Sun D.-W. Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review — Part II: Applications Innovative Food Science & Emerging Technologies 19 2013 15 28 10.1016/j.ifset.2013.04.016 Wythoff B.J. Backpropagation neural networks: A tutorial Chemometrics and Intelligent Laboratory Systems 18 2 1993 115 155 Zhang Y. Jiang H. Wang W. Feasibility of the detection of carrageenan adulteration in chicken meat using visible/near-infrared (vis/nir) hyperspectral imaging Applied Sciences 9 18 2019 3926 Zhao X. Wang W. Ni X. Chu X. Li Y.-F. Lu C. Utilising near-infrared hyperspectral imaging to detect low-level peanut powder contamination of whole wheat flour Biosystems Engineering 184 2019 55 68