Search

Hyperspectral imaging (HSI) is a promising technique to evaluate food quality in a rapid and non-destructive way. Several types of HSI sensors are now commercially available. The aim of the current study was to compare one single shot and two line-scanning based HSI sensors in different spectral regions by using two wavelength dispersive devices – a prism-grating-prism monochromator and a mosaic wavelength dispersion unit. The research sample was cured pork meat, which is a popular meat product due to its flavor and long shelf life. Three main chemical compositions of cured pork meat – moisture, protein and fat content – were evaluated. The spectral properties of the three sensors were employed to establish partial least squares regression (PLSR) predictive models by applying different optimal wavelengths selection strategies. The results showed that optimal wavelength selection was more important than the selection of the wavelength dispersion device. • Three hyperspectral imaging (HSI) systems were compared. • Prediction models for compositions of dry and wet cured pork were established. • Nine wavelengths were sufficient for predicting chemical compositions. • Slight differences between the two wavelengths dispersion systems were observed. • Optimal wavelength selection was a key step in improving models of HSI. [ABSTRACT FROM AUTHOR]

Abstract Secondary structure of proteins is closely related to the overall quality of meat. The sequence of the changes in chemical bonds caused by secondary structure modification and rapid detection method of secondary structure fractions was investigated by near-infrared hyperspectral imaging combined with generalized (G2D-CS) and heterospectral (H2D-CS) two-dimensional correlation analysis in this study. Synchronous and asynchronous spectral results from G2D-CS analysis indicated that, as α-helix fraction decreasing, the spectral intensity of N H stretching changed before that of the C H and O H, and the spectral fluctuation of O H stretching occurred after the carbonyl related C H band and before those aliphatic and aromatic C H band. Feature wavebands identified by H2D-CS analysis obtained good results with R2 CV of 0.836 in predicting α-helix fractions. The results of this study are useful for interpreting the secondary structure modification process of proteins during frozen storage and monitoring the secondary structure fraction in a rapid way. Highlights • NIR HSI was employed by the first time to monitor α-helix fraction of actomyosin. • G2D-CS was used for understanding the process of secondary structure modification. • H2D-CS was applied for identifying α-helix related NIR wavebands. • PLSR model built by the selected wavebands showed good results. [ABSTRACT FROM AUTHOR]

Abstract Protein is an important nutrient for people diet, and protein content is one of the most vital properties in quality assessment of pork meat. The potential of line-scanning based hyperspectral imaging (HSI) has been proved to be able to assess the protein content of meat, however quality of images obtained from line-scanning HSI is affected by sample transformation conditions. A novel single shot (snapshot) HSI sensor was employed to evaluate the protein content of numerous processed pork meat by using back propagation - neural network (BP-NN) and partial least squares regression (PLSR) predictive models. The results of BP-NN were better than that of PLSR. The best spectral profile was selected from reflectance, absorbance, and Kubelka–Munk (K–M) spectra by comparing the performances of their BP-NN models. The BP-NN model combined with absorbance spectra showed the best performance for evaluating the protein content of various processed pork meats with determination coefficient of cross-validation set (R2 CV) = 0.8318, and root mean square error of cross-validation set (RMSECV) = 8.38 mg/g, respectively. Results indicated the feasibility of determining the protein content of pork meats by means of single shot HSI. Highlights • The single shot hyperspectral imaging (HSI) could capture an image in 1/150 s. • Protein content of pork decreased after wet salting – dehydrating. • Absorbance spectra provided better accuracy. • BP-NN model was used to predict protein contents of pork by snapshot HSI. • Protein contents distribution of processed pork were showed by snapshot HSI. [ABSTRACT FROM AUTHOR]

Highlights • A multispectral model was built for DHA and EPA prediction. • A novel PN-GA algorithm was used to select the key variables. • MLR models showed better results for DHA and EPA in two fish fillets. • The dynamic changes of DHA and EPA were visualized. Abstract A multispectral model for the detection of docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) changes in grass carp and salmon fillet was developed using physarum network and genetic algorithm (PN-GA) method for the first time. Partial least-squares regression (PLSR), multiple linear regressions (MLR), and principal component regression (PCR) algorithms were used to predict the DHA and EPA using optimal wavelengths selected by PN-GA. The MLR models showed the best DHA prediction results for both grass carp and salmon fillets, and also showed good prediction for EPA in grass carp fillet but poor prediction in salmon fillet. The MLR models were then applied for visualizing the spatial distribution of DHA and EPA changes in two fish fillets. The current results demonstrated that a developed multispectral imaging system could be feasibly constructed for DHA and EPA measurement in fish species with the optimal wavelengths selected by PN-GA method. [ABSTRACT FROM AUTHOR]

This study aimed to simultaneously predict different textural features in grass carps fillets ( Ctenopharyngodon idella ) during vacuum freeze drying using hyperspectral imaging models developed based on integrated group wavelengths in the range of 400 nm–1000 nm. Warner-Bratzler shear force (WBSF) is commonly used as an objective indicator for tenderness evaluation in fish products. First, partial least squares regression (PLSR) was employed to develop a quantitative function between the textural variations and the spectra extracted from the acquired hyperspectral images in the full spectral range at mean and median. Then the important wavelengths most related to WBSF, hardness, gumminess and chewiness were separately selected by regression coefficients (RC) from PLSR. Moreover, the integrated wavelengths for all the textural characteristics were determined by RC by considering all the individual group wavelengths. Finally, PLSR was conducted using different simplified group wavelengths and the performance of all the simplified models at mean or median were compared. The results showed that the integrated group wavelengths at median spectra proved to be the best for simultaneous prediction of WBSF ( R P 2 = 0.8774), hardness ( R P 2 = 0.8523), gumminess ( R P 2 = 0.7982) and chewiness ( R P 2 = 0.8453). The current study should widen the applications of hyperspectral imaging in the food industry. [ABSTRACT FROM AUTHOR]

When considering food security and huge market interest, a high-efficiency method to ensure the authenticity of the food product is necessary. For this goal, spectral imaging was explored for quantitative detection of Irish organic wheat flour (OWF) adulterated with common wheat flour (WF), cassava flour (CaF) and corn flour (CoF). Hyperspectral images (900–1700 nm) of OWF samples with a series of adulteration percentages were collected. The acquired spectra were pre-processed by second derivative (2nd Der) and standard normal variate (SNV) before modelling. Then partial least squares regression (PLSR) and principal component regression (PCR) were employed for quantitative analysis of adulteration proportion of CoF, CaF and WF in OWF. To develop more effective simplified models, three groups of feature wavelengths were selected from the loading plots of principal component analysis (PCA), and first-derivative and mean centering iteration algorithm (FMCIA). The models developed using FMCIA were better than PCA. After, the corresponding feature wavelengths were further reduced based on model regression coefficients (RC). The optimal result of admixture detection was emerged by the RC-FMCIA-PLSR model, with a determination coefficient of prediction ( R 2 P ) of 0.973 and a root mean square error of prediction (RMSEP) of 0.036 for OWF adulterated with CoF, R 2 P of 0.986 and RMSEP of 0.026 for OWF adulterated with CaF, and R 2 P of 0.971 and RMSEP of 0.038 for OWF adulterated with WF. Visualization maps were generated by calculating the spectral response of each pixel on flour samples. This result indicates that spectral imaging integrated with multivariate analysis has the potential to authenticate the admixtures in specific wheat flour in the range of 3–75% (w/w). [ABSTRACT FROM AUTHOR]

Drying uniformity is one of the most important indicators in evaluating a drying technique as well as the final quality of dried products. In the current study, three drying approaches (hot-air drying (HAD), microwave-vacuum drying (MVD), and the combined method (HAD + MVD)) were applied to dehydrate mango slices. During the HAD + MVD process, the time required for hot-air drying was determined in terms of colour variations during hot-air drying. With the help of hyperspectral imaging in conjunction with multivariate data analysis and image processing, the moisture content distribution on mango slices subjected to different drying methods was visualised. Results showed a non-uniform drying property for mango slices dried by HAD or MVD individually, where HAD-dried samples had a higher moisture content in the centre but MVD-dried samples showed the opposite result. Drying uniformity was improved when HAD and MVD were combined, which produced dried products with an even moisture distribution. Mango slice samples dried by HAD + MVD showed a porous structure and a high percentage of colour retention. The current study led to the development of an effective combined HAD + MVD technique for enhancing drying uniformity for the industry. [ABSTRACT FROM AUTHOR]

This study investigated the feasibility of hyperspectral imaging (HSI) to predict moisture content (MC) values in heated - dehydrated (H-D) and cool - air - dehydrated (C-A-D) pork samples using the calibrated models established based on fresh (F), frozen - thawed (F-T), dry - salting - dehydrated (D-S-D) and wet - salting - dehydrated (W-S-D) pork samples. The full spectra were extracted from the region of interests (ROIs) in the spectral range of 400–1000 nm and the textural variables were extracted by gray-level gradient co-occurrence matrix (GLGCM) method from the first two PC images accounting for 98.73% of the total variance. Moreover, the optimal wavelengths were selected by regression coefficients (RC). Partial least-squares regression (PLSR) predictive model was developed based on the above extracted data and their mutual combination, in which the changes can be correlated with MC - related attributes. The results demonstrated that the PLSR model based on the incorporation of the optimal wavelengths and the textures (OW-T) from F, F-T, D- S -D and W-S-D samples were the best to predict MC in H-D and C-A-D samples with R p 2 of 0.9489 and RMSEP of 1.4736. Moreover, the generated visualization maps provided a rapid way to screen the MC values unequally distributed in the whole pork samples after diverse processing conditions. Therefore, it is feasible and promising to improve the applicability of the existing MC predictive models for MC prediction in the pork samples by supplementing more samples under different treatments as another prediction sets. [ABSTRACT FROM AUTHOR]

This study explored the potential of computer vision system (CVS) and hyperspectral imaging (HSI) technique covering spectral range of 900–1700 nm for identifying freezer burnt salmon fillets after frozen storage. Local binary pattern (LBP) descriptor was applied for the RGB image classification. Reflectance spectra were obtained from various positions surface and pretreated using the standard normal variate (SNV) transformation. TreeBagger classifier was used to build classification models for recognition and authentication of the freezer burnt flesh. The results suggested that hyperspectral discrimination performed much better than CVS with the correct classification rate (CCR) of 0.905 in validation and CCR of 0.945 in cross-validation. The effective wavelengths were selected based upon the feature importance in the TreeBagger model and the corresponding optimized model yielded CCR of 0.914 in validation and 0.978 in cross-validation. Overall, the outcome suggested the capability of HSI for rapid categorization of damaged regions on frozen salmon. [ABSTRACT FROM AUTHOR]

This study aimed to develop simplified models for rapid and nondestructive monitoring TVB-N contents during drying of cured meats based on a hyperspectral imaging system (HSI) in the spectral range of 400–1000 nm. Multivariate calibration models were first developed using partial least-squares regression (PLSR) and least-squares support vector machines (LS-SVM) in the full spectral range. In order to simplify the calibration model, a set of 9 feature wavelengths was then selected using the regression coefficient and two simplified models using PLSR and multiple linear regression (MLR) were subsequently established. The best simplified model obtained was based on MLR, with an R 2 p of 0.861 and RMSEP of 4.73. Mapping of TVB-N contents was realized by transferring the quantitative model to each pixel in the image to display protein degradation in cured pork slices at different drying periods. The results provided a possibility of realizing a multispectral imaging technique for on-line monitoring the TVB-N contents of cured meats during drying processes. [ABSTRACT FROM AUTHOR]

Biogenic amine index (BAI) is a sensitive indicator of meat freshness and quality. This study investigated the use of chemometric methods for analyzing hyperspectral imaging (HSI) data between 400 nm and 1000 nm to rapidly and non-destructively determine BAI values in pork. Partial least square regression (PLSR) model established using full wavelengths showed good results. In order to simplify the calibration model, four new PLSR and multiple linear regression (MLR) models based on the two sets of feature-related wavelengths selected by successive projections algorithm (SPA) and regression coefficients (RC) were built and compared. The optimized simplified model (RC-MLR) yielded excellent results with R 2 P of 0.957 and RMSEP of 4.866 mg/kg, which was thus used to visualize BAI value corresponding to each pixel of the image using pseudo color. In addition, the mechanisms of HSI for BAI determination were discussed. The established models used to determine BAI values were based on physiochemical changes associated with BAI generation in meat rather than direct detection of the BAI contents. The overall results of this study demonstrated that HSI data can be utilized to predict BAI values in pork based on chemometric analysis. [ABSTRACT FROM AUTHOR]

The freezing medium temperature and the freezing rate are two important parameters that affect the quality of frozen product. The traditional measurement of freezing parameters will destroy the integrity of the sample and can only be implemented during the freezing process. This study aimed to develop nondestructive hyperspectral imaging (HSI) methods to rapidly detect freezing parameters. The spectral features of the porcine meat samples in frozen state were studied, in which 90 pieces of porcine samples were frozen by different methods with different freezing medium (air and liquid) at different temperatures (from −20 to −120 °C) and freezing rates (from 0.307 to 5.1 cm/h). The result showed that the freezing process would strongly influence spectra of the frozen sample. The reflectance increased with the decrease in freezing medium temperatures, and the negative correlation reached a highly significant level. The freezing parameters did not change the position of the spectral peaks but altered the spectral intensity. Most changes were near 1070, 1172, 1420, 1586, and 1890 nm. The partial least-squares regression spectral models exhibited good performance for predicting freezing medium temperatures $$ \left({R}_c^2=0.898,{R}_p^2=0.844\right) $$ and freezing rates $$ \left({R}_c^2=0.879,{R}_p^2=0.829\right) $$ . The study confirmed that could be used for measuring freezing parameters of frozen product. This novel method will not damage the sample integrity, and measurement can be implemented anytime rather than only during the freezing process by traditional methods. [ABSTRACT FROM AUTHOR]

This study was conducted to assess the potential feasibility of using hyperspectral imaging (900–1700 nm) for rapid determination of the volatility of tuber compositions (VTC) and prediction of the tuber cooking degree (TCD) in low temperature baking (LTB). Tuber samples of six categories from different origins were imaged and calibrated. The partial least squares regression (PLSR) and three-layer back propagation artificial neural network (TBPANN) models were established to predict VTC and TCD using the entire spectral range and the feature wavelengths. The optimal combination of characteristic wavelengths (991, 1031, 1071, 1138, 1252, 1403, 1460 and 1641 nm) were selected by first derivative and mean centering iteration algorithm (FMCIA) rather than other conventional methods. Based on the qualified eight wavelengths, the FMCIA-TBPANN approach yielded greater overall performance for predicting both VTC and TCD. Furthermore, the distribution maps of VTC and TCD were generated using a resulting function to visualize each pixel of spectral image. This demonstrated the capability of spectral imaging technique for rapid and accurate evaluation of VTC and TCD during LTB. [ABSTRACT FROM AUTHOR]

The reliability and veracity of hyperspectral imaging integrated with multivariate analyses were investigated for authentication of sliced organic potato (OP) from non-organic tubers and rapid grading of tubers on the basis of different moisture levels. Hyperspectral images of all the tuber samples were obtained and their spectral data were extracted and pre-processed. Then, partial least squares discriminant analysis (PLSDA) model was established for recognition of the tested samples. Loading plots of the second derivative (SD) and principal component analysis (PCA) were used for selecting characteristic wavelengths. The OP samples were identified correctly (100% accuracy) from non-organic tubers by MC-PLSDA model using only characteristic wavelengths, with predicted sensitivities of 1.000, specificities of ⩾0.944, classification error of ⩽0.028, coefficient of determination ( R 2 P ) of no more than 0.979 and the root mean square error of prediction (RMSEP) of ⩽0.532 for each adulterate type. Another simplified PLSDA model was applied for grading tuber moisture levels, resulting in a correct classification of ⩾91.6%. The visualization results shown on classification maps achieved a rapid and convenient interpretation of tuber varieties and moisture levels. These results indicated that hyperspectral imaging has a great potential for discrimination of OP and identification of tuber moisture levels. [ABSTRACT FROM AUTHOR]

Spectral absorption index was proposed to extract the morphological features of the spectral curves in pork meat samples ( longissimus dorsi ) under the conditions including fresh, frozen–thawed, heated–dehydrated and brined–dehydrated. Savitzky–Golay (SG) smoothing and multiplicative scatter correction (MSC) were used for calibrating both the spectral reflectance and absorbance values. The absorption values were better than the reflectance values and the calibrated spectra by MSC were better than the raw and SG smoothing corrected spectra in building moisture content predictive models. The optimized partial least square regression (PLSR) model attained good results with the MSC calibrated spectral absorption values based on the spectral absorption index features ( R 2 P = 0.952, RMSEP = 1.396) and the optimal wavelengths selected by regression coefficients ( R 2 P = 0.966, RMSEP = 0.855), respectively. The models proved spectral absorption index was promising in spectral analysis to predict moisture content in pork samples using HSI techniques for the first time. [ABSTRACT FROM AUTHOR]

The prediction of moisture content uniformity on mango slices as affected by four different shapes (square, rectangle, regular triangle, and round shape) during microwave-vacuum drying (MVD) was investigated using near-infrared hyperspectral imaging in combination with multivariate chemometric analysis. Applying spectral pretreatment of a 2nd derivative followed by mean-center to raw spectra was found to be greatly beneficial for the reduction of noise and scattering levels. Seven wavelengths (951, 977, 1138, 1362, 1386, 1420, and 1440 nm) with larger absolute values of regression coefficients derived from a partial least square regression model were identified as feature variables for moisture prediction. An optimized model based on the selected wavelengths was developed using multivariate linear regression, achieving a high prediction accuracy with R p 2 = 0.993 and RMSEP = 1.282%. From the moisture distribution map, a similar non-uniform drying pattern was found on square, rectangle and regular triangle-shaped samples, while round-shaped mango slices achieved better drying results. Industrial relevance The current study suggested that NIR hyperspectral imaging was a promising technique in predicting the moisture content of mango slices during MVD, and non-uniformity of moisture distribution and the effect of sample geometry should be taken into account when the microwave-vacuum method is implemented in drying. [ABSTRACT FROM AUTHOR]

With consumer concerns increasing over food quality and safety, the food industry has begun to pay much more attention to the development of rapid and reliable food-evaluation systems over the years. As a result, there is a great need for manufacturers and retailers to operate effective real-time assessments for food quality and safety during food production and processing. Computer vision, comprising a nondestructive assessment approach, has the aptitude to estimate the characteristics of food products with its advantages of fast speed, ease of use, and minimal sample preparation. Specifically, computer vision systems are feasible for classifying food products into specific grades, detecting defects, and estimating properties such as color, shape, size, surface defects, and contamination. Therefore, in order to track the latest research developments of this technology in the agri-food industry, this review aims to present the fundamentals and instrumentation of computer vision systems with details of applications in quality assessment of agri-food products from 2007 to 2013 and also discuss its future trends in combination with spectroscopy. [ABSTRACT FROM PUBLISHER]

Mango slices were dried by microwave-vacuum drying using a domestic microwave oven equipped with a vacuum desiccator inside. Two lab-scale hyperspectral imaging (HSI) systems were employed for moisture prediction. The Page and the Two-term thin-layer drying models were suitable to describe the current drying process with a fitting goodness of R 2 = 0.978. Partial least square (PLS) was applied to correlate the mean spectrum of each slice and reference moisture content. With three waveband selection strategies, optimal wavebands corresponding to moisture prediction were identified. The best model RC-PLS-2 ( R p 2 = 0.972 and RMSEP = 4.611%) was implemented into the moisture visualization procedure. Moisture distribution map clearly showed that the moisture content in the central part of the mango slices was lower than that of other parts. The present study demonstrated that hyperspectral imaging was a useful tool for non-destructively and rapidly measuring and visualizing the moisture content during drying process. [ABSTRACT FROM AUTHOR]

Background Most contamination in agricultural and food products (AFP) is caused by undesirable microorganisms and how to detect microbial contaminants in a rapid and reliable way has always been a topic of great interest for researchers. By integrating traditional spectroscopy and digital imaging function into one system, hyperspectral imaging simultaneously providing spectral and spatial information is proposed and could be used as a smart and promising technology for microbial evaluation in AFP. Scope and approach In this review, we summarized the recent progresses and applications of hyperspectral imaging in replacing conventional time-consuming, tedious, labor-intensive and destructive techniques by providing a rapid, real-time, non-destructive and efficient alternative for detecting various microbial contaminants in raw and processed AFP. The great potential and importance of hyperspectral imaging is emphasized and approved with its satisfactory performance in microbial evaluation of various AFP. Key findings and conclusions Although hyperspectral imaging shows some obvious advantages, it still faces challenges in terms of accurate reference value measurement, fast data mining and advanced system development. With the further research and development of optics, it would be expected that efficient multispectral imaging systems would likely be developed and applied for online microbial evaluation in agricultural and food industry. [ABSTRACT FROM AUTHOR]

K value is an important freshness index widely used for indication of nucleotide degradation and assessment of chemical spoilage. The feasibility of hyperspectral imaging (400–1000 nm) for determination of K value in grass carp and silver carp fillets was investigated. Partial least square (PLS) regression and least square support vector machines (LS-SVM) models established using full wavelengths showed excellent performances and the PLS model was better with higher determination coefficients of prediction ( R 2 P = 0.936) and lower root mean square errors of prediction (RMSEP = 5.21%). The simplified PLS and LS-SVM models using the seven optimal wavelengths selected by successive projections algorithm (SPA) also presented good performances. The spatial distribution map of K value was generated by transferring the SPA-PLS model to each pixel of the images. The current study showed the suitability of using hyperspectral imaging to determine K value for evaluation of chemical spoilage and freshness of fish fillets. [ABSTRACT FROM AUTHOR]

As there is growing interest in process control for quality and safety in the meat industry, by integrating spectroscopy and imaging technologies into one system, hyperspectral imaging, or chemical or spectroscopic imaging has become an alternative analytical technique that can provide the spatial distribution of spectrum for fast and nondestructive detection of meat safety. This review addresses the configuration of the hyperspectral imaging system and safety indicators of muscle foods involving biological, chemical, and physical attributes and other associated hazards or poisons, which could cause safety problems. The emphasis focuses on applications of hyperspectral imaging techniques in the safety evaluation of muscle foods, including pork, beef, lamb, chicken, fish and other meat products. Although HSI can provide the spatial distribution of spectrum, characterized by overtones and combinations of the C–H, N–H, and O–H groups using different combinations of a light source, imaging spectrograph and camera, there still needs improvement to overcome the disadvantages of HSI technology for further applications at the industrial level. [ABSTRACT FROM AUTHOR]

• Shorter fibres in beef slices improved hot air dehydration efficiency. • Microwave vacuum drying showed higher rehydration rates in samples than hot air. • Beef slices showed higher shrinkage rates and expansion ratios than splits. • Principal component analysis could cluster samples by fibre and processing factors. • Hyperspectral imaging could visualise differences in sample processing behaviours. Effects of food microstructure and processing methods on moisture migration in food during processing are important for improving product quality. This study evaluated the effects of fibre directions and dehydration methods on beef dehydration and rehydration behaviours using near-infrared hyperspectral imaging. Beef slices (transversal cut) and splits (longitudinal cut) were dehydrated by hot air drying (HAD) and microwave vacuum drying (MVD). Results indicated the moisture in the slices evaporated faster than that in the splits during HAD and MVD processed samples exhibited higher rehydration rates than HAD ones. Principal component analysis showed great performance in clustering samples based on fibre directions, processing methods and time. The established simplified RC-SG-PLSR and RC-2nd Der-PLSR models possessed good prediction accuracies for moisture (R2 p = 0.994) and rehydration rate (R2 p = 0.889), respectively. Chemical distribution maps also confirmed the influence of fibre directions and dehydration methods on moisture migration in samples. [ABSTRACT FROM AUTHOR]

This study examined the potential of hyperspectral imaging (HSI) for rapid prediction of 2-thiobarbituric acid reactive substances (TBARS) content in chicken meat during refrigerated storage. Using the spectral data and the reference values of TBARS, a partial least square regression (PLSR) model was established and yielded acceptable results with regression coefficients in prediction ( R p ) of 0.944 and root mean squared errors estimated by prediction (RMSEP) of 0.081. To simplify the calibration model, ten optimal wavelengths were selected by successive projections algorithm (SPA). Then, a new SPA–PLSR model based on the selected wavelengths was built and showed good results with R p of 0.801 and RMSEP of 0.157. Finally, an image algorithm was developed to achieve image visualization of TBARS values in some representative samples. The encouraging results of this study demonstrated that HSI is suitable for determination of TBARS values for freshness evaluation in chicken meat. [ABSTRACT FROM AUTHOR]

Quality determination of frozen food is a time-consuming and laborious work as it normally takes a long time to thaw the frozen samples before measurements can be carried out. In this research, a rapid and non-destructive determination technique for frozen pork quality was tested with a hyperspectral imaging (HSI) system. In this study, 120 pieces of pork meat were frozen by four kinds of methods with various freezing temperatures from −20 to −120 °C. The hyperspectral images of the samples were acquired at the frozen state. Quality indicators including drip loss, pH value, color, cooking loss and Warner–Bratzler shear force (WBSF) of the samples were measured after thawing. The spectral characteristics of the frozen meat samples were studied and it was revealed that the reflectance at 1100 nm had a close relationship with the freezing temperature ( R =−0.832, p <0.01). Partial least squares regression (PLSR) was applied to establish the spectral models, and the models were then optimized. Results showed that the improved region of interest (ROI) method could be used to extract effective spectral information to withstand the interference of freezing, and choosing appropriate spectral bands and spectral pretreatment techniques were crucial to develop robust mathematical model. The performances of the models established were diverse based on different quality indicators. The coefficients of determination for prediction ( R p 2 ) for L *, cooking loss, b *, drip loss and a * were 0.907, 0.845, 0.814, 0.762, and 0.716, respectively. However there were low correlations ( R p 2 ) for pH and WBSF measurements. The current study indicated that HSI had the potential for non-destructive determination of frozen meat quality without thawing. [ABSTRACT FROM AUTHOR]

The feasibility of visible and near infrared hyperspectral imaging in the range of 400–1000 nm for determinating total viable counts (TVC) to evaluate microbial spoilage of fish fillets was investigated. Partial least square regression (PLSR) and least square support vector machines (LS-SVM) models established based on full wavelengths showed excellent performances and the LS-SVM model was better with higher residual predictive deviation (RPD) of 3.89, determination coefficients in prediction ( R 2 P ) of 0.93 and lower root mean square errors in prediction (RMSEP) of 0.49 log 10 CFU/g. Seven optimal wavelengths were selected by successive projections algorithm (SPA) and the simplified SPA-PLSR was better than SPA-LS-SVM models with RPD of 3.13, R 2 P of 0.90 and RMSEP of 0.57 log 10 CFU/g, and was transferred to each pixel of the hyperspectral images for generating the TVC distribution map. This study showed that hyperspectral imaging is suitable to determine TVC value for evaluating microbial spoilage of grass carp fillets in a rapid and non-invasive manner. [ABSTRACT FROM AUTHOR]

This study investigated the potential of hyperspectral imaging (HSI) for quantitative determination of total pigments in red meats, including beef, goose, and duck. Partial least squares regression (PLSR) was applied to correlate the spectral data with the reference values of total pigments measured by a traditional method. In order to simplify the PLSR model based on the full spectra, eleven optimal wavelengths were selected using successive projections algorithm (SPA). The new SPA-PLSR model yielded good results with the coefficient of determination ( R 2 p ) of 0.953, root mean square error (RMSEP) of 9.896, and ratio of prediction to deviation (RPD) of 4.628. Finally, distribution maps of total pigments in red meats were developed using an image processing algorithm. The overall results from this study indicated HSI had the capability for predicting total pigments in red meats. [ABSTRACT FROM AUTHOR]

In this study, the potential of hyperspectral imaging (HSI) for predicting hydroxyproline content in chicken meat was investigated. Spectral data contained in the hyperspectral images (400–1000 nm) of chicken meat was extracted, and a partial least square regression (PLSR) model was then developed for predicting hydroxyproline content. The model yielded acceptable results with regression coefficient in prediction ( R p ) of 0.874 and root mean error squares in prediction (RMESP) of 0.046. Based on the eight optimal wavelengths selected by regression coefficients (RC) from the PLSR model, a new RC-PLSR model was built and good results were shown with high R p of 0.854 and low RMSEP of 0.049. Finally, distribution maps of hydroxyproline were created by transferring the RC-PLSR model to each pixel in the hyperspectral images. The results demonstrated that HSI has the capability for rapid and non-destructive determination of hydroxyproline content in chicken meat. [ABSTRACT FROM AUTHOR]

The total counts of Enterobacteriaceae and Pseudomonas spp. (EPC) appeared on edible salmon flesh were determined by using near-infrared (NIR) (900–1700 nm) hyperspectral imaging. Three-dimensional hyperspectral images ( x , y , λ ) of salmon samples were acquired at different storage time. Spectra ( λ ) extracted in reflectance ( R ) unit and two other transformed spectra units, absorbance ( A ) and Kubelka–Munck (KM), were prepared to relate to the measured EPC data by using partial least square (PLS) regression. Based on the three spectra parameters, three full wavelength PLS models defined as FR-PLS, FA-PLS and FKM-PLS were developed with all correlation coefficients of prediction ( R P ) over 0.900. To simplify these models, wavelengths holding the most important information were selected by executing competitive adaptive reweighted sampling (CARS) algorithm. Better performance was found in the resulting simplified R-PLS model (defined as FR S -PLS model) which was established with only nine important wavelengths (931, 1138, 1175, 1242, 1359, 1628, 1641, 1652 and 1655 nm) selected from R spectra. The absolute difference between root mean square errors of calibration (RMSEC) and prediction (RMSEP) in the FR S -PLS model was 0.063, less than half (44%) of that of the original FR-PLS model. By applying the FR S -PLS model to the 2-D images ( x , y ), EPC distribution maps were generated to visualize the spatial variation of EPC and the adaptability of the FR S -PLS model for EPC evaluation was further demonstrated with these distribution maps in which different colors indicated different degrees of EPC contamination. To sum up, NIR hyperspectral imaging technology shows a great potential to predict the EPC contamination in salmon flesh. In view of the results obtained from this study, a multi-spectral imaging system could be developed and further refined for online detection applications in fish industry. [ABSTRACT FROM AUTHOR]

The suitability of hyperspectral imaging technique (400–1000 nm) was investigated to determine the thiobarbituric acid (TBA) value for monitoring lipid oxidation in fish fillets during cold storage at 4 °C for 0, 2, 5, and 8 days. The PLSR calibration model was established with full spectral region between the spectral data extracted from the hyperspectral images and the reference TBA values and showed good performance for predicting TBA value with determination coefficients ( R 2 P ) of 0.8325 and root-mean-square errors of prediction (RMSEP) of 0.1172 mg MDA/kg flesh. Two simplified PLSR and MLR models were built and compared using the selected ten most important wavelengths. The optimised MLR model yielded satisfactory results with R 2 P of 0.8395 and RMSEP of 0.1147 mg MDA/kg flesh, which was used to visualise the TBA values distribution in fish fillets. The whole results confirmed that using hyperspectral imaging technique as a rapid and non-destructive tool is suitable for the determination of TBA values for monitoring lipid oxidation and evaluation of fish freshness. [ABSTRACT FROM AUTHOR]

This study investigated the potential of visible and near infrared (Vis/NIR) hyperspectral imaging (HSI) to differentiate between free-range and broiler chicken meats. 120 hyperspectral images of chicken fillets were acquired and then calibrated for reflectance. Spectral data were extracted from the region of interest (ROI), followed by multiple scatter correction (MSC) to reduce the noise. Successive projection algorithm (SPA) was used to select optimal wavelengths from the full spectra. On the other hand, principal component analysis (PCA) was applied to select optimum characteristic images, and the first two principal component (PC) images were selected because PC1 and PC2 explained over 95% of variances of all spectra. Then, gray-level gradient co-occurrence matrix (GLGCM) was implemented on PC1 and PC2 images to extract 30 textural variables in total. Based on data fusion, classification models were established, in which the radial basis function-support vector machine (RBF-SVM) model gave the best results with high correct classification rate (CCR) of 93.33% for the prediction samples, demonstrating that combining spectra with texture data were effective for differentiating between free-range and broiler chicken meats. [ABSTRACT FROM AUTHOR]

The potential of visible and near infrared hyperspectral imaging was investigated as a rapid and nondestructive technique for classifying fresh and frozen-thawed meats by integrating critical spectral and image features extracted from hyperspectral images in the region of 400–1000 nm. Six feature wavelengths (400, 446, 477, 516, 592 and 686 nm) were identified using uninformative variable elimination and successive projections algorithm. Image textural features of the principal component images from hyperspectral images were obtained using histogram statistics (HS), gray level co-occurrence matrix (GLCM) and gray level-gradient co-occurrence matrix (GLGCM). By these spectral and textural features, probabilistic neural network (PNN) models for classification of fresh and frozen-thawed pork meats were established. Compared with the models using the optimum wavelengths only, optimum wavelengths with HS image features, and optimum wavelengths with GLCM image features, the model integrating optimum wavelengths with GLGCM gave the highest classification rate of 93.14% and 90.91% for calibration and validation sets, respectively. Results indicated that the classification accuracy can be improved by combining spectral features with textural features and the fusion of critical spectral and textural features had better potential than single spectral extraction in classifying fresh and frozen-thawed pork meat. [ABSTRACT FROM AUTHOR]

There are a large number of ham products produced and consumed throughout the world. The quality of ham is usually evaluated by human quality inspector based on grading standards. To maintain ham quality, processors need to carry out some destructive analysis of samples, which is usually laborious, time consuming and possesses the variability associated with human perception. Therefore, there is a need for the development of objective quality measurement or evaluation systems. Processors expect in-situ non-destructive measurements that can provide objective parameters with the incorporation of all the quality attributes and correlate them with the most demanding quality standards and consumer preferences. Hyperspectral imaging (HSI) is a promising technology that integrates conventional imaging and spectroscopy to attain both spatial and spectral information from the sample. Although the hyperspectral imaging approach has not been widely used for the quality evaluation of hams, it possesses a great potential in attaining information about different physico-chemical properties of ham. With the application of this technology, quality of ham could be ensured in a non-destructive manner and with confidence. The aim of this review is to discuss about the opportunities relating to hyperspectral imaging techniques in the quality evaluation and control of ham. [ABSTRACT FROM AUTHOR]

This study was conducted to investigate the potential of hyperspectral imaging technique in a rapid and non-invasive manner for measuring colour distribution of grass carp fillets during cold storage. The quantitative calibration models were established between the spectral data extracted from the hyperspectral images and the measured colour reference values by partial least squares regression (PLSR) and least squares support vector machines (LS-SVM). The performance of two spectral ranges of 400-1,000 and 1,000-2,500 nm was compared to select the best spectral range for further colour analysis of grass carp fillets. The LS-SVM model using the whole spectral range possessed better performance than the PLSR model for predicting colour components of L* and a* with higher coefficients of determination ( R) of 0.916 and 0.905 and lower root-mean-square errors of prediction (RMSEPs) of 2.876 and 2.253, respectively. Seven (466, 525, 590, 620, 715, 850 and 955 nm) and five (465, 585, 660, 720 and 950 nm) optimal wavelengths carrying the most important and sensitive information were recognized and selected using successive projections algorithm (SPA) for predicting L* and a*, with R values of 0.912 and 0.891 being obtained from the optimized SPA-LS-SVM models established based on the selected valuable wavelengths. In addition, the visualization maps of colour distribution of the examined fish fillets were acquired. The overall results of this study demonstrated that hyperspectral imaging technique in the spectral range of 400-1,000 nm has the potential to be used as an objective and promising tool for rapid and non-destructive measurement of colour distribution of grass carp fillets. [ABSTRACT FROM AUTHOR]

In this study, wavelet textural analysis was applied to hyperspectral images in the visible and near-infrared (VIS/NIR) region (400-1,000 nm) for differentiation between fresh and frozen-thawed pork. The spectral data of acquired hyperspectral images were analyzed using partial least squares (PLS) regression and five wavelengths (462, 488, 611, 629, and 678 nm) were selected as the feature wavelengths by the regression coefficients from the PLS model. The fourth-order daubechies wavelet ('db4') was used to serve as the wavelet mother function for wavelet textural extraction of the feature images at the above selected feature wavelengths with the wavelet decomposition level from 1 to 4. Four textural features were calculated in the horizontal, vertical, and diagonal orientations at each level. Forty-eight textural features were extracted from each feature image and used to differentiate between fresh and frozen-thawed pork samples by least-squares support vector machine (LS-SVM) model. Wavelet texture extracted from all five feature images at first decomposition level was identified as optimal wavelet texture combination, resulting in the highest classification accuracy for the LS-SVM models (98.48 % for the training set and 93.18 % for the testing set). Based on the texture combination, the quality attributes of pork meat could be predicted with correlation coefficients of calibration ( r) of 0.982 and 0.913, and correlation coefficients of prediction ( r) of 0.845 and 0.711 for pH and thawing loss, respectively. The results showed the possibility of developing a fast and reliable hyperspectral system for discrimination between fresh and frozen-thawed pork samples based on wavelet texture in the VIS/NIR wavelength range. [ABSTRACT FROM AUTHOR]

In this study, hyperspectral imaging working in near-infrared (NIR) region (900-1700nm) was applied to evaluate surface lactic acid bacteria (LAB) spoilage of farmed salmon flesh during cold storage. Hyperspectral images of salmon samples were acquired at different storage times. Spectral information within regions of interest (ROIs) of images were extracted to relate to reference LAB values measured by standard pour plate method. Least-squares support vector machine (LS-SVM) algorithm was used to calibrate the full NIR range spectral data, resulting in regression coefficients of prediction (RP) of 0.929 with root mean square error of prediction (RMSEP) of 0.515. Competitive adaptive reweighted sampling (CARS) algorithm was employed to reduce the spectral redundancy and identify the most informative wavelengths (MIWs) most related with LAB prediction across the whole wavelength range. Eight individual MIWs at 1155nm, 1255nm, 1373nm, 1376nm, 1436nm, 1641nm, 1665nm and 1689nm were finally selected from the full 239 wavelengths. Based on the selected MIWs, a new optimised model named CARS-LS-SVM was established, leading to RP of 0.925 with RMSEP of 0.531. At last, the CARS-LS-SVM model was transferred to each pixel of hyperspectral images of samples and colour maps were generated for visualising the LAB spoilage process in salmon flesh. The overall results indicated that NIR hyperspectral imaging is very potential and could be used as a rapid, non-destructive and efficient technique for LAB evaluation in salmon flesh. [ABSTRACT FROM AUTHOR]

Highlights: [•] We introduced recent advances in hyperspectral imaging (HSI) systems. [•] We reviewed applications of detecting quality attributes of red meats using HSI. [•] We discussed challenges in using HSI such as dimensionality reduction. [•] Improving hardware and exploiting new algorithms are future trends of HSI. [Copyright &y& Elsevier]

Highlights: [•] First time hyperspectral imaging was used to monitor meat salting process. [•] Spectral profiles were compared to improve model performance. [•] Moisture contents of porcine meat was predicted with good accuracy. [•] Optimised PLSR and MLR models were developed. [•] Chemical images were developed to display moisture evolution and migration. [Copyright &y& Elsevier]

Highlights: [•] First time, the texture distribution of salmon fillet was assessed non-invasively. [•] Both image texture and spectral features were extracted from hyperspectral images. [•] Spectra at 400–1000nm and 967–1634nm were compared for assess texture of salmon. [•] The selection of IOW and POW was proposed instead of traditional variable selection. [•] The distribution maps of TPA parameters were visualised in a pixel-wise manner. [ABSTRACT FROM AUTHOR]

During the past decade, hyperspectral imaging (HSI) has been rapidly developing and widely applied in the food industry by virtue of the use of chemometric techniques in which wavelength selection methods play an important role. This paper is a review of such variable selection methods and their limitations, describing the basic taxonomy of the methods and their respective advantages and disadvantages. Special attention is paid to recent developments in wavelength selection techniques for HSI in the field of food quality and safety evaluations. Typical and commonly used methods in HSI, such as partial least squares regression, stepwise regression and spectrum analysis, are described in detail. Some sophisticated methods, such as successive projections algorithm, uninformative variable elimination, simulated annealing, artificial neural network and genetic algorithm methods, are also discussed. Finally, new methods not currently used but that could have substantial impact on the field are presented. In short, this review provides an overview of wavelength selection methods in food-related areas and offers a thoughtful perspective on future potentials and challenges in the development of HSI systems. [ABSTRACT FROM AUTHOR]

Abstract: Hyperspectral imaging technique in the spectral wavelength range of 400–1000nm was implemented in this study to determine the total volatile basic nitrogen (TVB-N) contents of grass carp fillets during the frozen storage. The quantitative calibration models were built between the spectral data extracted from the hyperspectral images and the reference measured TVB-N values by using partial least squares regression (PLSR) and least squares support vector machines (LS-SVM). The LS-SVM model using full spectral range had a better performance than the PLSR model for prediction of TVB-N value with the corresponding coefficients of determination (R2 P) of 0.916 and 0.905, and root-mean-square errors of prediction (RMSEP) of 2.346% and 2.749%, respectively. Nine optimal wavelengths (420, 466, 523, 552, 595, 615, 717, 850 and 955nm) were selected using successive projections algorithm (SPA), and R2 P values of 0.902 and 0.891 with the corresponding RMSEP of 2.782% and 2.807% were obtained from the new optimized models established based on the selected valuable wavelengths. The best SPA-LS-SVM model was used to achieve the visualization map of TVB-N content distribution of the tested fish fillet samples. The results of this study indicated that hyperspectral imaging technique as an objective and promising tool is capable of determining TVB-N values for evaluation of fish freshness quality in a rapid and non-destructive way. Industrial relevance: The study showed that VIS–NIR hyperspectral imaging technique was an effective and powerful tool for rapid and non-destructive determination and assessment of fish fillet freshness for the fish industry. [Copyright &y& Elsevier]

Abstract: Water-holding capacity (WHC) is a primary quality determinant of salmon flesh. One of the limiting factors for not having a direct measurement of WHC for salmon quality grading is that current WHC measurements are destructive, time-consuming, and inefficient. In this study, two hyperspectral image systems operated in the visible and short-wave near infrared range (400–1000nm) and the long-wave near infrared range (897–1753nm) were applied for non-invasive determination of four WHC indices, namely percentage liquid loss (PLL), percentage water loss (PWL), percentage fat loss (PFL), and percentage water remained (PWR) of salmon flesh. Two calibration methods of partial least square regression (PLSR) and least-squares support vector machines (LS-SVM) were applied, respectively, to establish calibration models of WHC indices based on the spectral signatures of salmon flesh, and the performances of these two methods were compared to determine the optimal spectral calibration strategy. The performances were also compared between two hyperspectral image systems, when full range spectra were considered. Out of 121 wavelength variables, only thirteen (PLL), twelve (PWL), nine (PFL), and twelve variables (PWR) were selected as important variables by using competitive adaptive reweighted sampling (CARS) algorithm to reduce redundancy and collinearity of hyperspectral images. The CARS–PLSR combination was identified as the optimal method to calibrate the prediction models for WHC determination, resulting in good correlation coefficient of prediction ( ) of 0.941, 0.937, 0.815, and 0.970 for PLL, PWL, PFL, and PWR analysis, respectively. CARS–PLSR equations were obtained according to the regression coefficients of the CARS–PLSR models and were transferred to each pixel in the image for visualizing WHC indices in all portions of the salmon fillet. The overall results show that the laborious, time-consuming, and destructive traditional techniques could be replaced by hyperspectral imaging to provide a rapid and non-invasive measurement of WHC distribution in salmon flesh. [Copyright &y& Elsevier]

Abstract: Developing a rapid and non-destructive method for food safety and quality monitoring has become a crucial request from the meat industry. Hyperspectral imaging technique provides extraordinary advantages over the traditional imaging and spectroscopy techniques in food quality evaluation due to the spatial and spectral information that it can offer. In this study, a laboratory-based pushbroom hyperspectral imaging system in reflectance mode was developed in the near infrared (NIR) range (900–1700nm) for non-invasive determination of the major chemical compositions of beef. Beef samples collected from different breeds were scanned by the system followed by traditional assessment of their chemical composition by using the ordinary wet-chemical methods. The extracted spectral data and the measured quality parameters were modeled by partial least squares regression (PLSR) for predicting water, fat and protein contents yielding a reasonable accuracy with determination coefficients of 0.89, 0.84 and 0.86 concomitant with standard error of prediction (SEP) of 0.46%, 0.65% and 0.29%, respectively. Some image processing algorithms were developed and the most relevant wavelengths were selected to visualize the predicted chemical constituents in each pixel of the hyperspectral image yielding the spatially distributed visualizations of the sample contents. The results were promising and implied that hyperspectral imaging technique associated with appropriate chemometric multivariate analyses has a great potential for simultaneous assessment of various chemical constituents without using hazardous chemical reagents. [Copyright &y& Elsevier]

Abstract: In recent years, hyperspectral imaging has gained a wide recognition as a non-destructive and fast quality and safety analysis and assessment method for a wide range of food products. As the second part of this review, applications in quality and safety determination for food products are presented to illustrate the capability of this technique in the food industry for classification and grading, defect and disease detection, distribution visualization of chemical attributes, and evaluations of overall quality of meat, fish, fruits, vegetables, and other food products. The state of the art of hyperspectral imaging for each of the categories was summarized in the aspects of the investigated quality and safety attributes, the used systems (wavelength range, acquisition mode), the data analysis methods (feature extraction, multivariate calibration, variables selection), and the performance (correlation, error, visualization). With its success in different applications of food quality and safety analysis and assessment, it is evident that hyperspectral imaging can automate a variety of routine inspection tasks. Industrial relevance: It is anticipated that real-time food monitoring systems with this technique can be expected to meet the requirements of the modern industrial control and sorting systems in the near future. [Copyright &y& Elsevier]

Abstract: By integrating two classical optical sensing technologies of imaging and spectroscopy into one system, hyperspectral imaging can provide both spatial and spectral information, simultaneously. Therefore, hyperspectral imaging has the capability to rapidly and non-invasively monitor both physical and morphological characteristics and intrinsic chemical and molecular information of a food product in the purpose of quality and safety analysis and assessment. As the first part of this review, some fundamental knowledge about hyperspectral imaging is reviewed, which includes the relationship between spectroscopy, imaging, and hyperspectral imaging, principles of hyperspectral imaging, instruments for hyperspectral imaging, processing methods for data analysis, and discussion on advantages and disadvantages. Industrial relevance: It is anticipated that real-time food monitoring systems with this technique can be expected to meet the requirements of the modern industrial control and sorting systems in the near future. [Copyright &y& Elsevier]

Abstract: Identification of fresh and frozen–thawed meat is an important authenticity issue, although a challenging task. The potential application of a pushbroom hyperspectral imaging system in the near-infrared (NIR) range (900–1700nm) was explored as a rapid and non-destructive technique for the investigation of meat freshness. The effect of standard freeze–thaw routines was studied in pork samples from the longissimus dorsi muscle. Partial least squares discriminant analysis (PLS-DA) models were used to distinguish between fresh and frozen–thawed samples. Optimal wavelengths were selected and used for sample discrimination with reduced spectral data and image processing. Classification models with reduced spectral data achieved an overall correct classification of 100% for an independent set of samples. An image processing algorithm was also developed for visualizing the classification results. The best classification model obtained was successfully applied to the images to produce classification maps with high overall accuracy. Results confirmed the possibility of developing a fast and reliable system for discrimination between fresh and frozen–thawed pork based on reflectance in the NIR wavelength range. Industrial relevance: The application of NIR hyperspectral analysis could enable the development of a rapid, reliable and non-destructive method for authentication of fresh meat samples for the benefit of the retail sector and the consumer. [Copyright &y& Elsevier]

Food safety is a great public concern, and outbreaks of food-borne illnesses can lead to disturbance to the society. Consequently, fast and nondestructive methods are required for sensing the safety situation of produce. As an emerging technology, hyperspectral imaging has been successfully employed in food safety inspection and control. After presenting the fundamentals of hyperspectral imaging, this paper provides a comprehensive review on its application in determination of physical, chemical, and biological contamination on food products. Additionally, other studies, including detecting meat and meat bone in feedstuffs as well as organic residue on food processing equipment, are also reported due to their close relationship with food safety control. With these applications, it can be demonstrated that miscellaneous hyperspectral imaging techniques including near-infrared hyperspectral imaging, fluorescence hyperspectral imaging, and Raman hyperspectral imaging or their combinations are powerful tools for food safety surveillance. Moreover, it is envisaged that hyperspectral imaging can be considered as an alternative technique for conventional methods in realizing inspection automation, leading to the elimination of the occurrence of food safety problems at the utmost. [ABSTRACT FROM AUTHOR]

Abstract: This study was carried out for rapid and non-invasive measurement of color distribution in salmon fillet using hyperspectral imaging in long-wave near infrared spectral range (LW-NIR). Successive projections algorithm (SPA) was used to select effective wavelengths. Instead of selecting different sets of effective wavelengths for each color component respectively, instrumental effective wavelengths (IEWs) were identified for the prediction of all three color components, leading to reducing the number of band-pass filters for designing the multispectral imaging system. Meanwhile, predictive effective wavelengths (PEW) were further chosen from IEW to optimize calibration models. The final prediction models were considered as the multiple linear regression (MLR) models with PEW, which resulted in correlation coefficients (r C ) of 0.876, 0.744, and 0.803 for L*, a*, and b*, respectively. Finally, the visualization of color distribution in salmon fillets was achieved. The results demonstrated that LW-NIR hyperspectral imaging is a potential technique to quantitatively measure color distribution of salmon fillet in a rapid and non-invasive way. Industrial relevance: This study was conducted to evaluate the feasibility of a hyperspectral imaging system in the LW-NIR spectral region (964 to 1631 nm) for rapid measurement of color in intact salmon fillets. On the basis of the results and accompanying illustrations presented in this work, the study demonstrated the ability of the method based on LW-NIR hyperspectral imaging to measure color distribution of salmon fillets. This method was a rapid, contact-free, and consistent evaluation, and can be used as a reliable and rapid alternative to traditional colorimeter for measuring color of salmon fillet for the food industry. [Copyright &y& Elsevier]

Abstract: Hyperspectral imaging system operated in the near infrared (NIR) region (900–1700nm) was developed for non-contact measurement of surface colour, pH and tenderness of fresh beef. Hyperspectral images were acquired for beef samples and their spectral signatures were extracted. The real colour (expressed as L ∗ a ∗ b ∗), pH and tenderness of the same samples were recorded using traditional contact methods and then modelled with their corresponding spectral data using partial least square regression (PLSR). The L ∗, b ∗, pH and tenderness values were predicted with coefficients of determination () of 0.88, 0.81, 0.73 and 0.83 and root mean square errors estimated by cross validation (RMSECV) of 1.21, 0.57, 0.06 and 40.75, respectively. The weighted regression coefficients of the resulting PLSR models were used to identify the most important wavelengths and to reduce the high dimensionality of the hyperspectral data. By using these important wavelengths, image processing algorithm was developed to transfer the predicting models to every pixel in the image for visualizing colour and pH in all portions of the sample. The results demonstrated that NIR hyperspectral imaging system is a potential technique for non-destructive prediction of beef quality attributes, thus facilitating identification and classification of beef meat in a simple and fast way. With more improvement in terms of speed and processing, the hyperspectral imaging system could be effectively implemented in commercial meat product processing plants for non-destructive and rapid quality measurements. [Copyright &y& Elsevier]

Abstract: This study was carried out for post-mortem non-destructive prediction of water holding capacity (WHC) in fresh beef using near infrared (NIR) hyperspectral imaging. Hyperspectral images were acquired for different beef samples originated from different breeds and different muscles and their spectral signatures were extracted. Both principal component analysis (PCA) and partial least squares regression (PLSR) models were developed to obtain an overview of the systematic spectral variations and to correlate spectral data of beef samples to its real WHC estimated by drip loss method. Partial least squares modeling resulted in a coefficient of determination (R CV 2) of 0.89 and standard error estimated by cross validation (SECV) of 0.26%. The PLSR loadings showed that there are some important absorption peaks throughout the whole spectral range that had the greatest influence on the predictive models. Six wavelengths (940, 997, 1144, 1214, 1342, and 1443nm) were then chosen as important wavelengths to build a new PLS prediction model. The new model led to a coefficient of determination (R CV 2) of 0.87 and standard error estimated by cross validation (SECV) of 0.28%. Image processing algorithm was then developed to transfer the predicting model to each pixel in the image for visualizing drip loss in all portions of the sample. The results showed that hyperspectral imaging has the potential to predict drip loss non-destructively in a reasonable accuracy and the results could be visualised for identification and classification of beef muscles in a simple way. In addition to realize the difference in WHC within one sample, it was possible to accentuate the difference in samples having different drip loss values. [Copyright &y& Elsevier]

• Pork samples were processed with 5 methods under 3 conditions. • Hyperspectral imaging (HSI) was applied to evaluate processed pork samples. • Monounsaturated and Polyunsaturated fatty acids in samples were predicted by HSI. • LS-SVM models were better than PLSR models. Freezing, heating, and pickling are common processes for pork meats. Unsaturated fatty acids including monounsaturated fatty acids and polyunsaturated fatty acids are indispensable nutrition beneficial to human's health and growth. However, Unsaturated fatty acids are affected by processing methods. Hyperspectral imaging is a novel technique widely used for food quality and safety evaluation. In the current study, the contents of monounsaturated and polyunsaturated fatty acids were assessed by Hyperspectral imaging. Optimal wavelengths were selected by the regression coefficients curves of partial least squares regression models. The least-squares support vector machine models established achieved a better coefficient of determination in the Monte Carlo validation set than the partial least squares regression models developed and the R2 MV values for the least squares - support vector machine models based on selected optimal wavelengths were higher than 0.81. Finally, colour maps of the contents of monounsaturated and polyunsaturated fatty acids were developed. [ABSTRACT FROM AUTHOR]