Feature wavelength selection and model development for rapid determination of myoglobin content in nitrite-cured mutton using hyperspectral imaging

Monitoring of veterinary drug residues in mutton based on hyperspectral combined with explainable AI: A case study of OFX

Veterinary drug residues in meat seriously harm human health. Rapid and accurate detection of veterinary drug residues is necessary to minimize contamination. Taking ofloxacin (OFX) residues in mutton as an example, the near-infrared hyperspectral imaging combined with explainable AI was used to evaluate the importance of feature wavelengths in the convolutional neural network-stacked sparse auto-encoder (CNN-SSAE) model for chemical properties. Based on this, the qualitative (residue identification-residue level identification) and quantitative detection of OFX residues in mutton was realized. The results showed that the accuracy of CNN-SSAE in identifying residue and residue level of OFX was 100% and 93.65%, respectively, and the correlation coefficients for validation (R2P) in quantitative detection of OFX residue was 0.8980. In addition, SHapley Additive exPlanation (SHAP) values were used to identify feature wavelengths that contribute the most in the CNN-SSAE model, which effectively explained the quality attribute information that spectral and chemical values may improve the predicted results in the model decision process. The reliability of the CNN-SSAE model was evaluated by statistical validation methods (F-test and T-test). Finally, the visualization diagram of OFX content distribution was established. This study provides a method reference for explainability detection of veterinary drug residues.

Non-Destructive Detection of Chilled Mutton Freshness Using a Dual-Branch Hierarchical Spectral Feature-Aware Network

Precise detection of meat freshness levels is essential for food consumer safety and real-time quality monitoring. This study aims to achieve the high-accuracy freshness detection of chilled mutton freshness by integrating hyperspectral imaging with deep learning methods. Although hyperspectral data can effectively capture changes in mutton freshness, sparse raw spectra require optimal data processing strategies to minimize redundancy. Therefore, this study employs a multi-stage data processing approach to enhance the purity of feature spectra. Meanwhile, to address issues such as overlapping feature categories, imbalanced sample distributions, and insufficient intermediate features, we propose a Dual-Branch Hierarchical Spectral Feature-Aware Network (DBHSNet) for chilled mutton freshness detection. First, at the feature interaction stage, the PBCA module addresses the drawback that global and local branches in a conventional dual-branch framework tend to perceive spectral features independently. By enabling effective information exchange and bidirectional flow between the two branches, and injecting positional information into each spectral band, the model's awareness of sequential spectral bands is enhanced. Second, at the feature fusion stage, the task-driven MSMHA module is introduced to address the dynamics of freshness variation and the accumulation of different metabolites. By leveraging multi-head attention and cross-scale fusion, the model more effectively captures both the overall spectral variation trends and fine-grained feature details. Third, at the classification output stage, dynamic loss weighting is set according to training epochs and relative losses to balance classification performance, effectively mitigating the impact of insufficiently discriminative intermediate features. The results demonstrate that the DBHSNet enables a more precise assessment of mutton freshness, achieving up to 7.59% higher accuracy than conventional methods under the same preprocessing conditions, while maintaining superior weighted metrics. Overall, this study offers a novel approach for mutton freshness detection and provides valuable support for freshness monitoring in cold-chain meat systems.

Correlation between quality change and hydrogen sulfide in aquatic product: Detection of hydrogen sulfide and its potential applications using bigeye tuna (Thunnus obesus) model during cold storage

Hydrogen sulfide (H2S) is an metabolic product of tuna during the spoilage, and relationship between H2S and tuna quality has not been specifically studied. This study detected changes in H2S content, H2S precursor substances, and related enzymes based on the formation pathway of H2S. H2S content increased of tuna resulted in significant increases in contents of cystathionine β-synthase, cystathionine γ-lyase, 3-mercapto pyruvate sulfotransferase, cysteine aminotransferase and methionine, while content of cysteine decreased which provided H2S formation. Cysteine and methionine metabolism, sulfur metabolism and histidine metabolism were metabolic pathways to assess H2S accumulation. Canonical correlation analysis showed that H2S content was significantly correlated with total volatile base nitrogen, total viable count (p < 0.05). This study elucidates the universality of H2S as an index for assessing seafood quality, utilizing quality indicators and modeling. Our findings offer a theoretical foundation and potential practical applications for improving the quality control of aquatic products.

Advancements in Hyperspectral Imaging for Assessing Nutritional Parameters in Muscle Food: Current Research and Future Trends

Assessing the nutritional value of muscle food (MF) necessitates comprehensive component analysis. Traditional chemical analytical methods are often time-intensive, destructive, and environmentally detrimental, requiring specialized laboratory expertise. Hyperspectral imaging (HSI) emerges as an innovative technique that effectively integrates spectral and spatial information to enable rapid, nondestructive, and multidimensional predictions of nutritional parameters in MF. This Review examines the cutting-edge advancements in HSI technology, elucidating its novel technical and methodological dimensions. It systematically explores the principles and methodologies of HSI, presenting recent research and diverse applications in predicting MF nutritional parameters, and evaluates HSI's significant advantages and current limitations while addressing field-specific challenges and prospective research trends, ultimately positioning HSI as a potentially transformative tool in ensuring meat industry quality and safety.

Early diagnosis of Cladosporium fulvum in greenhouse tomato plants based on visible/near-infrared (VIS/NIR) and near-infrared (NIR) data fusion

Plant diseases can inflict varying degrees of damage on agricultural production. Therefore, identifying a rapid, non-destructive early diagnostic method is crucial for safeguarding plants. Cladosporium fulvum (C. fulvum) is one of the major diseases in tomato growth. This work presents a method of data fusion using two hyperspectral imaging systems of visible/near-infrared (VIS/NIR) and near-infrared (NIR) spectroscopy for the early diagnosis of C. fulvum in greenhouse tomatoes. First, hyperspectral images of samples at health and different times of infection were collected. The average spectral data of the image regions of interest were extracted and preprocessed for subsequent spectral datasets. Then different classification models were established for VIS/NIR and NIR data, optimized through various variable selection and data fusion methods. The principal component analysis-radial basis function neural network (PCA-RBF) model established using low-level data fusion achieved optimal results, achieving accuracies of 100% and 99.3% for calibration and prediction, respectively. Moreover, both the macro-averaged F1 (Macro-F1) values reached 1, and the geometric mean (G-mean) values reached 1 and 1, respectively. The results indicated that it was feasible to establish a PCA-RBF model by using the hyperspectral technique with low-level data fusion for the early detection of C. fulvum in greenhouse tomatoes.

Study on Rapid Non-Destructive Detection Method of Corn Freshness Based on Hyperspectral Imaging Technology

(1) Background: To achieve the rapid, non-destructive detection of corn freshness and staleness for better use in the storage, processing and utilization of corn. (2) Methods: In this study, three varieties of corn were subjected to accelerated aging treatment to study the trend in fatty acid values of corn. The study focused on the use of hyperspectral imaging technology to collect information from corn samples with different aging levels. Spectral data were preprocessed by a convolutional smoothing derivative method (SG, SG1, SG2), derivative method (D1, D2), multiple scattering correction (MSC), and standard normal transform (SNV); the characteristic wavelengths were extracted by the competitive adaptive reweighting method (CARS) and successive projection algorithm (SPA); a neural network (BP) and random forest (RF) were utilized to establish a prediction model for the quantification of fatty acid values of corn. And, the distribution of fatty acid values was visualized based on fatty acid values under the corresponding optimal prediction model. (3) Results: With the prolongation of the aging time, all three varieties of corn showed an overall increasing trend. The fatty acid value of corn can be used as the most important index for characterizing the degree of aging of corn. SG2-SPA-RF was the quantitative prediction model for optimal fatty acid values of corn. The model extracted 31 wavelengths, only 12.11% of the total number of wavelengths, where the coefficient of determination RP2 of the test set was 0.9655 and the root mean square error (RMSE) was 3.6255. (4) Conclusions: This study can provide a reliable and effective new method for the rapid non-destructive testing of corn freshness.

A new comprehensive quantitative index for the assessment of essential amino acid quality in beef using Vis-NIR hyperspectral imaging combined with LSTM

The quality of beef is usually predicted by measuring a single index rather than a comprehensive index. To precisely determine the essential amino acid (EAA) contents in 360 beef samples, the feasibility of optimized spectral detection techniques based on the comprehensive EAA index (CEI) and comprehensive weight index (CWI) constructed by factor analysis was explored. Two-dimensional correlation spectroscopy (2D-COS) was used to analyse the mechanisms of spectral peak shifts in complex disturbance systems with CEI and CWI contents, and 15 sensitive feature variables were extracted to establish a quantitative analysis model of a long short-term memory network (LSTM). The results indicated that 2D-COS had good predictive performance in both CEI-LSTM (R2P of 0.9095 and RPD of 2.76) and CWI-LSTM (R2P of 0.8449 and RPD of 2.45), which reduced data information by 88%. This indicates that utilizing 2D-COS can eliminate collinearity and redundant information among variables while achieving data dimensionality reduction and simplification of calibration models. Furthermore, a spatial distribution map of the comprehensive EAA content was generated by combining the optimal prediction model. This study demonstrated that the comprehensive index method furnishes a new approach to rapidly evaluate EAA content.

Discrimination among Fresh, Frozen-Stored and Frozen-Thawed Beef Cuts by Hyperspectral Imaging

The detection of the storage state of frozen meat, especially meat frozen-thawed several times, has always been important for food safety inspections. Hyperspectral imaging (HSI) is widely applied to detect the freshness and quality of meat or meat products. This study investigated the feasibility of the low-cost HSI system, combined with the chemometrics method, to classify beef cuts among fresh (F), frozen-stored (F-S), frozen-thawed three times (F-T-3) and frozen-thawed five times (F-T-5). A compact, low-cost HSI system was designed and calibrated for beef sample measurement. The classification model was developed for meat analysis with a method to distinguish fat and muscle, a CARS algorithm to extract the optimal wavelength subset and three classifiers to identify each beef cut among different freezing processes. The results demonstrated that classification models based on feature variables extracted from differentiated tissue spectra achieved better performances, with ACCs of 92.75% for PLS-DA, 97.83% for SVM and 95.03% for BP-ANN. A visualization map was proposed to provide detailed information about the changes in freshness of beef cuts after freeze-thawing. Furthermore, this study demonstrated the potential of implementing a reasonably priced HSI system in the food industry.

Predictive Analysis of Linoleic Acid in Red Meat Employing Advanced Ensemble Models of Bayesian and CNN-Bi-LSTM Decision Layer Fusion Based Hyperspectral Imaging

Rapid non-destructive testing technologies are effectively used to analyze and evaluate the linoleic acid content while processing fresh meat products. In current study, hyperspectral imaging (HSI) technology was combined with deep learning optimization algorithm to model and analyze the linoleic acid content in 252 mixed red meat samples. A comparative study was conducted by experimenting mixed sample data preprocessing methods and feature wavelength extraction methods depending on the distribution of linoleic acid content. Initially, convolutional neural network Bi-directional long short-term memory (CNN-Bi-LSTM) model was constructed to reduce the loss of the fully connected layer extracted feature information and optimize the prediction effect. In addition, the prediction process of overfitting phenomenon in the CNN-Bi-LSTM model was also targeted. The Bayesian-CNN-Bi-LSTM (Bayes-CNN-Bi-LSTM) model was proposed to improve the linoleic acid prediction in red meat through iterative optimization of Gaussian process acceleration function. Results showed that best preprocessing effect was achieved by using the detrending algorithm, while 11 feature wavelengths extracted by variable combination population analysis (VCPA) method effectively contained characteristic group information of linoleic acid. The Bi-directional LSTM (Bi-LSTM) model combined with the feature extraction data set of VCPA method predicted 0.860 Rp2 value of linoleic acid content in red meat. The CNN-Bi-LSTM model achieved an Rp2 of 0.889, and the optimized Bayes-CNN-Bi-LSTM model was constructed to achieve the best prediction with an Rp2 of 0.909. This study provided a reference for the rapid synchronous detection of mixed sample indicators, and a theoretical basis for the development of hyperspectral on-line detection equipment.

Discrimination of wheat flour grade based on PSO-SVM of hyperspectral technique

Rapid detection of wheat flour grade played an important role in the food industry. In this work, hyperspectral technology was used to detect five types of wheat flour. An analysis model was established based on the reflectance of samples at 968 ∼ 2576 nm. Moreover, multivariate scattering correction (MSC), standard normalized variate (SNV), and Savitzky-Golay (S-G) convolution smoothing were used for preprocessing, which was employed to reduce the influence of noise in the original spectrum. In order to simplify the model, competing adaptive reweighted sampling (CARS), successive projection algorithm (SPA), uninformative variable elimination (UVE) and the UVE-CARS algorithm were applied to extract feature wavelengths. Both partial least squares discriminant analysis (PLS-DA) model and support vector machine (SVM) model were established according to feature wavelengths. Furthermore, particle swarm optimization (PSO) algorithm was adopted to optimize the search of SVM model parameters, such as the penalty coefficient c and the regularization coefficient g. Experimental results suggested that the non-linear discriminant model for wheat flour grades was better than the linear discriminant model. It was considered that the MSC-UVE-CARS-PSO-SVM model achieved the best forecasting results for wheat flour grade discrimination, with 100% accuracy both in the calibration set and the validation set. It further shows that the classification of wheat flour grade can be effectively realized by using the hyperspectral and SVM discriminant analysis model, which proves the potential of hyperspectral reflectance technology in the qualitative analysis of wheat flour grade.

The Identification of Fritillaria Species Using Hyperspectral Imaging with Enhanced One-Dimensional Convolutional Neural Networks via Attention Mechanism

Combining deep learning and hyperspectral imaging (HSI) has proven to be an effective approach in the quality control of medicinal and edible plants. Nonetheless, hyperspectral data contains redundant information and highly correlated characteristic bands, which can adversely impact sample identification. To address this issue, we proposed an enhanced one-dimensional convolutional neural network (1DCNN) with an attention mechanism. Given an intermediate feature map, two attention modules are constructed along two separate dimensions, channel and spectral, and then combined to enhance relevant features and to suppress irrelevant ones. Validated by Fritillaria datasets, the results demonstrate that an attention-enhanced 1DCNN model outperforms several machine learning algorithms and shows consistent improvements over a vanilla 1DCNN. Notably under VNIR and SWIR lenses, the model obtained 98.97% and 99.35% for binary classification between Fritillariae Cirrhosae Bulbus (FCB) and other non-FCB species, respectively. Additionally, it still achieved an extraordinary accuracy of 97.64% and 98.39% for eight-category classification among Fritillaria species. This study demonstrated the application of HSI with artificial intelligence can serve as a reliable, efficient, and non-destructive quality control method for authenticating Fritillaria species. Moreover, our findings also illustrated the great potential of the attention mechanism in enhancing the performance of the vanilla 1DCNN method, providing reference for other HSI-related quality controls of plants with medicinal and edible uses.

The mechanism of Leuconostoc mesenteroides subsp. IMAU:80679 in improving meat color: Myoglobin oxidation inhibition and myoglobin derivatives formation based on multi enzyme-like activities

The Leuconostoc mesenteroides subsp. IMAU:80679 (LM) was chosen for its superior capability in enhancing redness, and was incubated in a broth system containing metmyoglobin (MetMb) to investigate its mechanisms for color improvement. The a* value of LM group reached its highest level of 52.75 ± 1.04 at 24 h, significantly higher than control of 19.75 ± 0.6 (p < 0.05). The addition of LM could inhibit myoglobin oxidation to some extent. Meanwhile, higher content of nitrosylmyoglobin (NOMb) and Zn-protoporphyrin (Znpp) were observed in LM samples during the whole incubation period. Furthermore, enzymatic activity and encoded genes related to MetMb reduction and pigment formation were determined to explain its possible mechanism on color enhancement. Finally, by extracting crude enzymes and adding them to meat batters, the redness of crude enzyme group was comparable to that achieved with 20 ppm nitrite, providing a potential method on compensating for nitrite/nitrate substitution in meat products.

A Combination of Near-Infrared Hyperspectral Imaging with Two-Dimensional Correlation Analysis for Monitoring the Content of Alanine in Beef

Alanine (Ala), as the most important free amino acid, plays a significant role in food taste characteristics and human health regulation. The feasibility of using near-infrared hyperspectral imaging (NIR-HSI) combined with two-dimensional correlation spectroscopy (2D-COS) analysis to predict beef Ala content quickly and nondestructively is first proposed in this study. With Ala content as the external disturbance condition, the sequence of chemical bond changes caused by synchronous and asynchronous correlation spectrum changes in 2D-COS was analyzed, and local sensitive variables closely related to Ala content were obtained. On this basis, the simplified linear, nonlinear, and artificial neural network models developed by the weighted coefficient based on the feature wavelength extraction method were compared. The results show that with the change in Ala content in beef, the double-frequency absorption of the C-H bond of CH₂ in the chemical bond sequence occurred prior to the third vibration of the C=O bond and the first stretching of O-H in COOH. Furthermore, the wavelength within the 1136-1478 nm spectrum range was obtained as the local study area of Ala content. The linear partial least squares regression (PLSR) model based on effective wavelengths was selected by competitive adaptive reweighted sampling (CARS) from 2D-COS analysis, and provided excellent results (R²_C of 0.8141, R²_P of 0.8458, and RPDp of 2.54). Finally, the visual distribution of Ala content in beef was produced by the optimal simplified combination model. The results show that 2D-COS combined with NIR-HSI could be used as an effective method to monitor Ala content in beef.

Quantitative study of impact damage on yellow peaches based on reflectance, absorbance and Kubelka-Munk spectral data

Impact damage is one of the main forms of damage during the postharvest transportation and processing of yellow peaches. Thus, a quantitative prediction of the impact damage degree of yellow peaches is significant for their postharvest grading. In the present study, mechanical parameters such as the damage area, absorbed energy and maximum force were obtained based on a single pendulum collision device and an intelligent data acquisition system. The reflection spectra (R) of the damaged areas of yellow peaches were collected by a hyperspectral imaging system and transformed into absorbance (A) spectra and Kubelka-Munk (K-M) spectra. The R, A and K-M spectra were preprocessed by standard normal variables (SNV), moving average (MA) and Gaussian filtering (GF). Partial least squares regression (PLSR) models and support vector regression (SVR) models based on original and preprocessed spectra were established, respectively. By comparative analysis, the spectral data with better prediction performance (raw or preprocessed spectra) were selected from all spectra, and the characteristic wavelengths were selected by competitive adaptive reweighted sampling (CARS) and uninformative variable elimination (UVE). The PLSR and SVR models based on characteristic wavelengths were established. The results revealed that the prediction performance of the K-M-GF-CARS-PLSR model is the best. For the damage area, absorbed energy and maximum force, the R _P ² and RMSEP of the K-M-GF-CARS-PLSR model were 0.870 and 77.865 mm², 0.772 and 1.065 J, 0.895 and 47.996 N, respectively. Furthermore, the values of their RPD were 2.700, 1.768 and 3.050, respectively. The characteristic wavelengths of the model were 18.8%, 10.2% and 21.6%, respectively. The results of this study showed that there was a strong correlation between the mechanical parameters and K-M spectrum, which demonstrates the feasibility of quantitatively predicting the damage degree of yellow peaches based on the K-M spectrum. Therefore, the results of this work not only provide theoretical guidance for the postharvest grading of fruits, but also enrich the theoretical system of biomechanics.

Estimation of amino acid contents in maize leaves based on hyperspectral imaging

Estimation of the amino acid content in maize leaves is helpful for improving maize yield estimation and nitrogen use efficiency. Hyperspectral imaging can be used to obtain the physiological and biochemical parameters of maize leaves with the advantages of being rapid, non-destructive, and high throughput. This study aims to estimate the multiple amino acid contents in maize leaves using hyperspectral imaging data. Two nitrogen (N) fertilizer experiments were carried out to obtain the hyperspectral images of fresh maize leaves. The partial least squares regression (PLSR) method was used to build the estimation models of various amino acid contents by using the reflectance of all bands, sensitive band range, and sensitive bands. The models were then validated with the independent dataset. The results showed that (1) the spectral reflectance of most amino acids was more sensitive in the range of 400-717.08 nm than other bands. The estimation accuracy was better by using the reflectance of the sensitive band range than that of all bands; (2) the sensitive bands of most amino acids were in the ranges of 505.39-605 nm and 651-714 nm; and (3) among the 24 amino acids, the estimation models of the β-aminobutyric acid, ornithine, citrulline, methionine, and histidine achieved higher accuracy than those of other amino acids, with the R ², relative root mean square error (RE), and relative percent deviation (RPD) of the measured and estimated value of testing samples in the range of 0.84-0.96, 8.79%-19.77%, and 2.58-5.18, respectively. This study can provide a non-destructive and rapid diagnostic method for genetic sensitive analysis and variety improvement of maize.

Deep generative neural networks for spectral image processing

An artificial intelligence approach based on deep generative neural networks for spectral imaging processing was proposed. The key idea was to treat different spectral image processing operations such as segmentation, regression, and classification as image-to-image translation tasks. For the image-to-image translation, the conditional generative adversarial networks were used. As a baseline comparison, the traditional chemometric approach based on pixels wise modelling was demonstrated. The analysis was presented with two real data sets related to fruit property prediction and kernel and shell classification of walnuts. The presented artificial intelligence approach for spectral image processing can provide benefits for any field of science where spectral imaging and processing is widely performed.

Research on nondestructive identification of grape varieties based on EEMD-DWT and hyperspectral image

Grape varieties are directly related to the quality and sales price of table grapes and consumed products (raisin, wine, grape juice, etc.). To satisfy the identification requirements of rapid, accurate, and nondestructive detection, an improved denoising algorithm based on ensemble empirical mode decomposition (EEMD) and discrete wavelet transform (DWT) is proposed to couple with the hyperspectral image (HSI) of grape varieties in this study. First, the hyperspectral data of grape varieties are collected by using HSI instrument, and denoised by the proposed EEMD-DWT and other denoising algorithms. CARS-SPA (competitive adaptive reweighed sampling coupled with successive projections algorithm) is introduced to select the effective wavelengths and a discriminative model is constructed by using support vector machine (SVM). Finally, Monte Carlo experiments verified that EEMD-DWT was an effective and powerful spectra denoising method, and the SVM model constructed by combining with CARS-SPA had an excellent identification accuracy (99.3125%). The results suggested that the key wavelengths selected by using CARS-SPA and EEMD-DWT could be an alternative to the deal with HSI, and its potential to become a method for identifying grape varieties. PRACTICAL APPLICATION: Traditional grape varieties identification methods are destructive and time consuming. Therefore, HSI technology is applied to realize fast and nondestructive identification of grape varieties in this study. The research results indicate that it is feasible to combine HSI technology with machine learning algorithm to discriminate grape varieties. It is of great significance for grape grading and the promotion of excellent varieties, and also provides reference for grape industry producers to identify grape varieties quickly and accurately.

Development of a novel quantitative function between spectral value and metmyoglobin content in Tan mutton

This study was aimed to establish a quantitative function between spectral reflectance values and metmyoglobin (MetMb) content in Tan mutton during refrigeration. Near-infrared hyperspectral data combined with generalized two-dimensional correlation spectroscopy (G2D-COS) method to identify characteristic bands and investigate the sequence of chemical waveband changes. Characteristic wavebands identified by G2D-COS analysis had the best performance in predicting the content of MetMb, with a high R²p of 0.849, a low RMSEP of 2.695 and a high RPD of 2.786. The results showed that the G2D-COS may be a powerful tool for describing intensity changes of MetMb band. The partial least square regression method was used to develop the relationships between the spectral values and MetMb content in Tan mutton meat for predicting MetMb content. This study has provided a convenient and rapid non-destructive quantitative method for assessing the color of Tan mutton meat.