Current Articles
The effect of calcium sequestering salts on the quality of milk-pea blended protein processed cheese
2024, 39(5)
:1-8.
doi: 10.12187/2024.05.001
Abstract:
Milk protein and pea protein were used as raw materials to produce milk-pea blended protein processed cheese. This study investigated the effect of three types of calcium sequestering salts (trisodium citrate (TSC), dibasic sodium phosphate (DSP), and sodium hexametaphosphate (SHMP)) at different mass fractions (0.5%, 1.0%, 1.5%, and 2.0%) on the quality of the blended protein processed cheese based on functional properties, texture properties, sensory evaluation, and microstructure. The results showed that milk protein and pea protein could substitute for natural cheese in processed cheese formulations. As the mass fraction of the calcium sequestering salts increased, the meltability of the blended protein processed cheese demonstrated a consistent upward trend, accompanied by a reduction in fat separation, and the TSC had shown good effects in improving product melting characteristics and inhibiting oil precipitation. When the mass fraction of TSC was 1.0%, the hardness of the blended protein processed cheese was lower, whereas its total sensory score, elasticity, cohesiveness, and adhesiveness were higher. Meanwhile, casein network micelles were arranged in a regular pattern, with fat globules and minerals uniformly distributed throughout the matrix. Consequently, the incorporation of TSC could improve the quality of milk-pea blended protein processed cheese by enhancing its emulsification,stability and other functional characteristics.
Milk protein and pea protein were used as raw materials to produce milk-pea blended protein processed cheese. This study investigated the effect of three types of calcium sequestering salts (trisodium citrate (TSC), dibasic sodium phosphate (DSP), and sodium hexametaphosphate (SHMP)) at different mass fractions (0.5%, 1.0%, 1.5%, and 2.0%) on the quality of the blended protein processed cheese based on functional properties, texture properties, sensory evaluation, and microstructure. The results showed that milk protein and pea protein could substitute for natural cheese in processed cheese formulations. As the mass fraction of the calcium sequestering salts increased, the meltability of the blended protein processed cheese demonstrated a consistent upward trend, accompanied by a reduction in fat separation, and the TSC had shown good effects in improving product melting characteristics and inhibiting oil precipitation. When the mass fraction of TSC was 1.0%, the hardness of the blended protein processed cheese was lower, whereas its total sensory score, elasticity, cohesiveness, and adhesiveness were higher. Meanwhile, casein network micelles were arranged in a regular pattern, with fat globules and minerals uniformly distributed throughout the matrix. Consequently, the incorporation of TSC could improve the quality of milk-pea blended protein processed cheese by enhancing its emulsification,stability and other functional characteristics.
2024, 39(5)
:9-17.
doi: 10.12187/2024.05.002
Abstract:
Using black tea and peanut as the primary raw materials, fat-free milk powder and white granulated sugar as auxiliary materials, a black tea-peanut protein compound beverage with rich nutrition and good taste was prepared. The study aimed to analyze the sensory score, protein content, fat content and osmotic pressure of the black tea-peanut protein compound beverage to study the best technological conditions and nutritional characteristics. The results showed that the optimal process conditions of the black tea-peanut protein compound beverage were as follows: the volume ratio between peanut protein pulp to black tea juice was 1∶2, the addition of fat-free milk powder was 10%, and the addition of white granulated sugar was 2%. Under these conditions, the prepared compound beverage had the highest sensory score (86 points). The protein content was 5.51 g/100 g, fat content was 5.90 g/100 g, total phenolic content was 139.04 μg/mL, total flavonoid content was 258.98 μg/mL, the fat droplets were evenly distributed and had a smaller mean particle size (7.54 μm), the osmotic pressure (336.00 mmol/kg) was close to the human blood osmotic pressure, which could meet consumers' pursuit of sensory quality and health diet.
Using black tea and peanut as the primary raw materials, fat-free milk powder and white granulated sugar as auxiliary materials, a black tea-peanut protein compound beverage with rich nutrition and good taste was prepared. The study aimed to analyze the sensory score, protein content, fat content and osmotic pressure of the black tea-peanut protein compound beverage to study the best technological conditions and nutritional characteristics. The results showed that the optimal process conditions of the black tea-peanut protein compound beverage were as follows: the volume ratio between peanut protein pulp to black tea juice was 1∶2, the addition of fat-free milk powder was 10%, and the addition of white granulated sugar was 2%. Under these conditions, the prepared compound beverage had the highest sensory score (86 points). The protein content was 5.51 g/100 g, fat content was 5.90 g/100 g, total phenolic content was 139.04 μg/mL, total flavonoid content was 258.98 μg/mL, the fat droplets were evenly distributed and had a smaller mean particle size (7.54 μm), the osmotic pressure (336.00 mmol/kg) was close to the human blood osmotic pressure, which could meet consumers' pursuit of sensory quality and health diet.
2024, 39(5)
:18-28.
doi: 10.12187/2024.05.003
Abstract:
Soybean protein isolate (SPI), pea protein (PP) and chickpea protein (CP) were used as the main base materials, and wheat gluten (WG) was used as the auxiliary material, to study the effects of different plant protein compounds and ratios on the quality properties (sensory score, textural properties, texturization and microstructure) of plant-based meat. The results showed that the addition of WG could affect the sensory score of plant-based meat. For SPI plant-based meat, when the mass fraction of WG was 40%, the sensory score of plant-based meat was the highest ((68.68±0.85) points). For the SPI/WG system, the addition of PP increased the hardness of plant-based meat, while the addition of CP reduced the texturization of plant-based meat. For PP plant-based meat, when the mass fraction of WG was 10%, the sensory score of plant-based meat was the highest ((75.80±0.76) points). In the PP/WG system, as the SPI or CP addition increased, the tissue degree of plant-based meat decreased. For CP plant-based meat, when the mass fraction of WG was 30%, the sensory score of plant-based meat was the highest ((73.60±0.42) points). For the CP/WG system, the addition of PP or SPI could significantly reduce the hardness of plant-based meat. In the ternary protein compound plant-based meat, the PP/CP/WG (80∶10∶10) blend system had the highest sensory score ((77.30±0.57) points), hardness ((25.78±0.61) kg), and chewiness ((4.66±0.16) kg). Therefore, compounding different plant proteins was an effective method to improve the fiber structure and organization of plant-based meat. Therefore, compounding different plant proteins was an effective method to enhance the fibrous structure and texturization of plant-based meat.
Soybean protein isolate (SPI), pea protein (PP) and chickpea protein (CP) were used as the main base materials, and wheat gluten (WG) was used as the auxiliary material, to study the effects of different plant protein compounds and ratios on the quality properties (sensory score, textural properties, texturization and microstructure) of plant-based meat. The results showed that the addition of WG could affect the sensory score of plant-based meat. For SPI plant-based meat, when the mass fraction of WG was 40%, the sensory score of plant-based meat was the highest ((68.68±0.85) points). For the SPI/WG system, the addition of PP increased the hardness of plant-based meat, while the addition of CP reduced the texturization of plant-based meat. For PP plant-based meat, when the mass fraction of WG was 10%, the sensory score of plant-based meat was the highest ((75.80±0.76) points). In the PP/WG system, as the SPI or CP addition increased, the tissue degree of plant-based meat decreased. For CP plant-based meat, when the mass fraction of WG was 30%, the sensory score of plant-based meat was the highest ((73.60±0.42) points). For the CP/WG system, the addition of PP or SPI could significantly reduce the hardness of plant-based meat. In the ternary protein compound plant-based meat, the PP/CP/WG (80∶10∶10) blend system had the highest sensory score ((77.30±0.57) points), hardness ((25.78±0.61) kg), and chewiness ((4.66±0.16) kg). Therefore, compounding different plant proteins was an effective method to improve the fiber structure and organization of plant-based meat. Therefore, compounding different plant proteins was an effective method to enhance the fibrous structure and texturization of plant-based meat.
2024, 39(5)
:29-39.
doi: 10.12187/2024.05.004
Abstract:
The wheat Angiotensin Converting Enzyme (ACE) inhibitory peptide was prepared by ultrasonic-assisted enzymatic hydrolysis method, and the inhibitory rate of ACE was the main index, and the degree of hydrolysis of gluten powder was the secondary index. The preparation process was optimized by single factor test and response surface method, and the stability of the inhibitory peptides was studied. The results showed that alkaline protease was suitable for enzymatic hydrolysis of gluten powder to produce wheat ACE inhibitory peptides. The optimal preparation conditions were ultrasonic time of 17 min, ultrasonic power of 300 W, enzymatic hydrolysis temperature of 60 ℃, enzymatic hydrolysis time of 2.7 h, enzyme dosage of 3600 U/g, and gluten powder mass fraction of 5.1%. Under these conditions, the ACE inhibition rate of the prepared wheat ACE inhibitory peptides was 72.90%, and the hydrophobic amino acid content was 29.37 g/100 g. When the relative molecular weight of the inhibitory peptides was less than 3 kDa, it had good environmental stability and thermal stability, and also had good stability in a certain concentration of K+ and Mg2+, and could still maintain 79.26% of the original activity after simulated digestion in vitro. Therefore, ultrasound-assisted enzymatic hydrolysis method is an effective method to prepare wheat ACE inhibitory peptides.
The wheat Angiotensin Converting Enzyme (ACE) inhibitory peptide was prepared by ultrasonic-assisted enzymatic hydrolysis method, and the inhibitory rate of ACE was the main index, and the degree of hydrolysis of gluten powder was the secondary index. The preparation process was optimized by single factor test and response surface method, and the stability of the inhibitory peptides was studied. The results showed that alkaline protease was suitable for enzymatic hydrolysis of gluten powder to produce wheat ACE inhibitory peptides. The optimal preparation conditions were ultrasonic time of 17 min, ultrasonic power of 300 W, enzymatic hydrolysis temperature of 60 ℃, enzymatic hydrolysis time of 2.7 h, enzyme dosage of 3600 U/g, and gluten powder mass fraction of 5.1%. Under these conditions, the ACE inhibition rate of the prepared wheat ACE inhibitory peptides was 72.90%, and the hydrophobic amino acid content was 29.37 g/100 g. When the relative molecular weight of the inhibitory peptides was less than 3 kDa, it had good environmental stability and thermal stability, and also had good stability in a certain concentration of K+ and Mg2+, and could still maintain 79.26% of the original activity after simulated digestion in vitro. Therefore, ultrasound-assisted enzymatic hydrolysis method is an effective method to prepare wheat ACE inhibitory peptides.
2024, 39(5)
:40-49.
doi: 10.12187/2024.05.005
Abstract:
In order to expand the application range of chickpea protein isolate (CPI) and improve the stability of low oil emulsion systems, this study used CPI as an emulsifier in oil-in-water (O/W) type CPI emulsions and investigated the effect of ultrasound treatment (20 kHz, 450 W) with different durations (0 min, 3 min, 6 min, 9 min, and 12 min) on the stability of CPI emulsions. The results showed that ultrasound treatment could significantly improve the stability of CPI emulsions compared with the untreated CPI emulsions. At 12 min of ultrasound treatment, the emulsification activity index (EAI) and emulsion stability index (ESI) were increased to (51.67±0.12) m2/g and (99.32±0.13) min, respectively. The average particle size became the smallest ((3.52±0.25) μm), resulting in finer and more uniform oil droplets. The absolute value of zeta-potential was the largest (52.57±1.31 mV), the Turbiscan stability index (TSI) was the lowest, and the storage, thermal and freeze-thaw stability of the CPI emulsions were all improved. As revealed by cold-field electron microscopy, CPI emulsions treated by ultrasound had a dense and regular internal structure, with smaller droplets and less spacing. In conclusion, a certain duration ultrasound treatment could improve the stability of CPI prepared emulsions, and the better effect was achieved at 12 min of ultrasound treatment.
In order to expand the application range of chickpea protein isolate (CPI) and improve the stability of low oil emulsion systems, this study used CPI as an emulsifier in oil-in-water (O/W) type CPI emulsions and investigated the effect of ultrasound treatment (20 kHz, 450 W) with different durations (0 min, 3 min, 6 min, 9 min, and 12 min) on the stability of CPI emulsions. The results showed that ultrasound treatment could significantly improve the stability of CPI emulsions compared with the untreated CPI emulsions. At 12 min of ultrasound treatment, the emulsification activity index (EAI) and emulsion stability index (ESI) were increased to (51.67±0.12) m2/g and (99.32±0.13) min, respectively. The average particle size became the smallest ((3.52±0.25) μm), resulting in finer and more uniform oil droplets. The absolute value of zeta-potential was the largest (52.57±1.31 mV), the Turbiscan stability index (TSI) was the lowest, and the storage, thermal and freeze-thaw stability of the CPI emulsions were all improved. As revealed by cold-field electron microscopy, CPI emulsions treated by ultrasound had a dense and regular internal structure, with smaller droplets and less spacing. In conclusion, a certain duration ultrasound treatment could improve the stability of CPI prepared emulsions, and the better effect was achieved at 12 min of ultrasound treatment.
2024, 39(5)
:50-59.
doi: 10.12187/2024.05.006
Abstract:
An overview of data fusion strategies based on spectroscopy, mass spectrometry, chromatography and other detection technologies and their current research and application in authenticity identification of edible oils was presented, pointing out that: at present, detection technologies widely used for authenticity identification of edible oils including spectroscopy, chromatography, mass spectrometry and electronic sensors. However, a single detection technique often focused only on a specific data or index, which could not fully eliminate the superposition effect, baseline drift and noise when the ingredients contained in edible oils were more complex. Data fusion strategies were categorized into three types: data layer fusion, feature layer fusion and decision layer fusion. Combined with chemometrics methods, the data obtained by different detection technologies could be integrated to obtain and extract richer data feature information, thus improving the authenticity identification of edible oils. Data fusion between various novel detection technologies, or between new and traditional spectroscopy, mass spectrometry, chromatography and other detection technologies, which could quickly and accurately achieved the identification of adulteration of edible oils, variety classification and origin traceability. In the future, in-depth research could be carried out on the improvement of the existing analytical methods, the development of new fusion algorithms combined with deep learning algorithms, and the introduction of cloud computing to improve real-time edible oil identification, so as to promote the development of data fusion strategy in the field of edible oil authenticity identification.
An overview of data fusion strategies based on spectroscopy, mass spectrometry, chromatography and other detection technologies and their current research and application in authenticity identification of edible oils was presented, pointing out that: at present, detection technologies widely used for authenticity identification of edible oils including spectroscopy, chromatography, mass spectrometry and electronic sensors. However, a single detection technique often focused only on a specific data or index, which could not fully eliminate the superposition effect, baseline drift and noise when the ingredients contained in edible oils were more complex. Data fusion strategies were categorized into three types: data layer fusion, feature layer fusion and decision layer fusion. Combined with chemometrics methods, the data obtained by different detection technologies could be integrated to obtain and extract richer data feature information, thus improving the authenticity identification of edible oils. Data fusion between various novel detection technologies, or between new and traditional spectroscopy, mass spectrometry, chromatography and other detection technologies, which could quickly and accurately achieved the identification of adulteration of edible oils, variety classification and origin traceability. In the future, in-depth research could be carried out on the improvement of the existing analytical methods, the development of new fusion algorithms combined with deep learning algorithms, and the introduction of cloud computing to improve real-time edible oil identification, so as to promote the development of data fusion strategy in the field of edible oil authenticity identification.
2024, 39(5)
:60-70.
doi: 10.12187/2024.05.007
Abstract:
A modified QuEChERS method using magnetic graphene oxide as purification material was developed to determine four heterocyclic aromatic amines (IQ, Harman, Norharman and Phe-P-1) in ham sausage by ultra-high performance liquid chromatography (UPLC). Analytical performance of this method was evaluated through methodological validation and applied to the detection of real samples. The results showed that compared with traditional purification materials, the prepared magnetic graphene oxide displayed considerable abilities of matrix purification and achieved satisfactory HAAs recoveries, which could be used as the purification material in the modified QuEChERS method. The optimal conditions for the modified QuEChERS method were as follows: alkaline acetonitrile (containing ammonium hydroxide with volume fraction of 1%) was used as the extraction solvent, the volume of water was 1 mL, the vortex extraction time was 5 min, the combination of anhydrous sodium sulfate/sodium chloride was chosen as the extraction salt with the dosage of 0.50 g/0.13 g, the centrifugation time was 3 min, the dosage of magnetic graphene oxide was 7.5 mg and the purification time was 1 min. Under these conditions, all the correlation coefficients (R2) for the targeted HAAs were greater than 0.9995. The recoveries of four HAAs at three spiked levels (low, medium and high) were 99.9%~104.3%, 101.2%~106.7% and 85.7%~101.2%, respectively. The limit of detection (LOD) was 0.71~1.52 ng/g. The limit of quantification (LOQ) ranged from 1.00 ng/g to 2.53 ng/g. The intra-day and inter-day precisions were less than 4.4% and 4.9% with good accuracy and precision. The four HAAs were not detected in any of the tested ham sausages. However, some unknown substances were found and their contents were higher than the LOQ of the established QuEChERS method. Compared with other commonly used sample pretreatment techniques, this method had advantages of speed, simplicity, low cost per sample and less organic solvents, which was suitable for the determination of HAAs in ham sausage.
A modified QuEChERS method using magnetic graphene oxide as purification material was developed to determine four heterocyclic aromatic amines (IQ, Harman, Norharman and Phe-P-1) in ham sausage by ultra-high performance liquid chromatography (UPLC). Analytical performance of this method was evaluated through methodological validation and applied to the detection of real samples. The results showed that compared with traditional purification materials, the prepared magnetic graphene oxide displayed considerable abilities of matrix purification and achieved satisfactory HAAs recoveries, which could be used as the purification material in the modified QuEChERS method. The optimal conditions for the modified QuEChERS method were as follows: alkaline acetonitrile (containing ammonium hydroxide with volume fraction of 1%) was used as the extraction solvent, the volume of water was 1 mL, the vortex extraction time was 5 min, the combination of anhydrous sodium sulfate/sodium chloride was chosen as the extraction salt with the dosage of 0.50 g/0.13 g, the centrifugation time was 3 min, the dosage of magnetic graphene oxide was 7.5 mg and the purification time was 1 min. Under these conditions, all the correlation coefficients (R2) for the targeted HAAs were greater than 0.9995. The recoveries of four HAAs at three spiked levels (low, medium and high) were 99.9%~104.3%, 101.2%~106.7% and 85.7%~101.2%, respectively. The limit of detection (LOD) was 0.71~1.52 ng/g. The limit of quantification (LOQ) ranged from 1.00 ng/g to 2.53 ng/g. The intra-day and inter-day precisions were less than 4.4% and 4.9% with good accuracy and precision. The four HAAs were not detected in any of the tested ham sausages. However, some unknown substances were found and their contents were higher than the LOQ of the established QuEChERS method. Compared with other commonly used sample pretreatment techniques, this method had advantages of speed, simplicity, low cost per sample and less organic solvents, which was suitable for the determination of HAAs in ham sausage.
2024, 39(5)
:71-77.
doi: 10.12187/2024.05.008
Abstract:
Addressing the current reliance on manual subjective judgment and strong subjectivity for assessing the cleanliness of tobacco conveyor belts, a cleanliness classification model (ResNet24_SC_Block) for tobacco conveyor belt using an improved ResNet was proposed. This model uses ResNet for classification with a network depth of 24 layers. SE and CBAM attention mechanisms were introduced into the residual module to improve the model's ability to capture features such as conveyor belt color and adhesion smoke scale. Using the tobacco leaf conveyor belt dataset to experiment with this model, the experimental results showed that the average values of Accuracy, Precision, Recall and F1 of the improved ResNet24_SC_Block model were 98.8%, 98.8% and 98.4%, respectively, which were 3.3%～3.8% higher than those of ResNet18 model and ResNet34 model. Compared with classic and newer networks such as GoogLeNet model and RegNet model, it improves by 2.1% to 17.9%. And the number of model parameters was reduced by 31.6% compared with ResNet34 model. This approach offered notable advantages in accurately and efficiently assessing the cleanliness level of tobacco conveyor belts, making it highly consequential and practically valuable for intelligent upgrades in cigarette manufacturing plants.
Addressing the current reliance on manual subjective judgment and strong subjectivity for assessing the cleanliness of tobacco conveyor belts, a cleanliness classification model (ResNet24_SC_Block) for tobacco conveyor belt using an improved ResNet was proposed. This model uses ResNet for classification with a network depth of 24 layers. SE and CBAM attention mechanisms were introduced into the residual module to improve the model's ability to capture features such as conveyor belt color and adhesion smoke scale. Using the tobacco leaf conveyor belt dataset to experiment with this model, the experimental results showed that the average values of Accuracy, Precision, Recall and F1 of the improved ResNet24_SC_Block model were 98.8%, 98.8% and 98.4%, respectively, which were 3.3%～3.8% higher than those of ResNet18 model and ResNet34 model. Compared with classic and newer networks such as GoogLeNet model and RegNet model, it improves by 2.1% to 17.9%. And the number of model parameters was reduced by 31.6% compared with ResNet34 model. This approach offered notable advantages in accurately and efficiently assessing the cleanliness level of tobacco conveyor belts, making it highly consequential and practically valuable for intelligent upgrades in cigarette manufacturing plants.
2024, 39(5)
:78-85.
doi: 10.12187/2024.05.009
Abstract:
To achieve accurate identification and efficient segmentation of adhesive tobacco shreds, a method for adhesive tobacco shreds recognition based on an improved Mask R-CNN (Mask Region-based Convolutional Neural Network) model was proposed. Firstly, adhesive tobacco shreds images were collected, and the dataset was augmented through image enhancement operations to expand it to the required sample size for training the model. Secondly, edge feature extraction and fractal feature extraction were performed on the adhesive tobacco shreds images in the training set based on Mask R-CNN, resulting in clearer and more continuous image edge features and texture feature information. Subsequently, the original features, edge features, and fractal features were fused to fully utilize features at different levels and enrich low-level features. Finally, by introducing a hybrid attention mechanism that focused on both channel and spatial dimensions of feature maps, the efficiency and accuracy of tobacco shred recognition were improved. Experimental results showed that the mean intersectionover union (Avg.MIoU) of the recognition method based on the improved Mask R-CNN model was 85.29%, and the mean class pixel accuracy (Avg.MPA) was 84.33%, under different adhesion conditions enabling precise identification of tobacco shreds and outperforming the original Mask R-CNN and DeepLabV3+ models. This method could rapidly and accurately identify and segment adhesive tobacco shreds, providing technical support for subsequent tobacco shred width detection.
To achieve accurate identification and efficient segmentation of adhesive tobacco shreds, a method for adhesive tobacco shreds recognition based on an improved Mask R-CNN (Mask Region-based Convolutional Neural Network) model was proposed. Firstly, adhesive tobacco shreds images were collected, and the dataset was augmented through image enhancement operations to expand it to the required sample size for training the model. Secondly, edge feature extraction and fractal feature extraction were performed on the adhesive tobacco shreds images in the training set based on Mask R-CNN, resulting in clearer and more continuous image edge features and texture feature information. Subsequently, the original features, edge features, and fractal features were fused to fully utilize features at different levels and enrich low-level features. Finally, by introducing a hybrid attention mechanism that focused on both channel and spatial dimensions of feature maps, the efficiency and accuracy of tobacco shred recognition were improved. Experimental results showed that the mean intersectionover union (Avg.MIoU) of the recognition method based on the improved Mask R-CNN model was 85.29%, and the mean class pixel accuracy (Avg.MPA) was 84.33%, under different adhesion conditions enabling precise identification of tobacco shreds and outperforming the original Mask R-CNN and DeepLabV3+ models. This method could rapidly and accurately identify and segment adhesive tobacco shreds, providing technical support for subsequent tobacco shred width detection.
2024, 39(5)
:86-94.
doi: 10.12187/2024.05.010
Abstract:
To address the challenge of non-destructive detection of sucrose solution application in the tobacco leaf processing stage, a discrimination model for sucrose solution application based on hyperspectral imaging and machine learning had been developed. Hyperspectral data of tobacco leaf samples with varying sucrose solution applications were first acquired using a visible-shortwave infrared hyperspectral imaging system and preprocessed with standard normal variate (SNV). Four discrimination models for sucrose solution application were then constructed and validated using full-spectrum data and principal component analysis (PCA) reduced data, in conjunction with support vector machine (SVM), logistic regression (LR), multilayer perceptron (MLP), and random forest (RF). The results showed that SNV preprocessing significantly enhanced the feature concentration of the hyperspectral data. When modeling with full-spectrum data, the models in the shortwave infrared band demonstrated significantly higher prediction accuracy compared to those in the visible light band, with the LR model in the shortwave infrared band achieving the highest accuracy of 98.23%. Compared to full-spectrum data modeling, the prediction accuracy of models using the top 10 principal components from PCA reduced data showed little change in the shortwave infrared band, while the RF model's accuracy in the visible light band improved significantly to 71.43%. In the visible light band, the highest accuracy for PCA-reduced data models corresponded to 217, 55, 47, and 59 principal components, while in the shortwave infrared band, the numbers were 13, 11, 117, and 46, respectively. Overall, LR and RF models exhibited superior predictive perf ormance, with the LR model based on PCA-reduced data in the shortwave infrared band maintaining high accuracy with fewer principal components, demonstrating the capability for rapid, non-destructive, and precise determination of sucrose solution application on tobacco leaves.
To address the challenge of non-destructive detection of sucrose solution application in the tobacco leaf processing stage, a discrimination model for sucrose solution application based on hyperspectral imaging and machine learning had been developed. Hyperspectral data of tobacco leaf samples with varying sucrose solution applications were first acquired using a visible-shortwave infrared hyperspectral imaging system and preprocessed with standard normal variate (SNV). Four discrimination models for sucrose solution application were then constructed and validated using full-spectrum data and principal component analysis (PCA) reduced data, in conjunction with support vector machine (SVM), logistic regression (LR), multilayer perceptron (MLP), and random forest (RF). The results showed that SNV preprocessing significantly enhanced the feature concentration of the hyperspectral data. When modeling with full-spectrum data, the models in the shortwave infrared band demonstrated significantly higher prediction accuracy compared to those in the visible light band, with the LR model in the shortwave infrared band achieving the highest accuracy of 98.23%. Compared to full-spectrum data modeling, the prediction accuracy of models using the top 10 principal components from PCA reduced data showed little change in the shortwave infrared band, while the RF model's accuracy in the visible light band improved significantly to 71.43%. In the visible light band, the highest accuracy for PCA-reduced data models corresponded to 217, 55, 47, and 59 principal components, while in the shortwave infrared band, the numbers were 13, 11, 117, and 46, respectively. Overall, LR and RF models exhibited superior predictive perf ormance, with the LR model based on PCA-reduced data in the shortwave infrared band maintaining high accuracy with fewer principal components, demonstrating the capability for rapid, non-destructive, and precise determination of sucrose solution application on tobacco leaves.
2024, 39(5)
:95-101.
doi: 10.12187/2024.05.011
Abstract:
To address the issues of complexity, low accuracy, high personnel requirements, and inability to adapt to the demand for large-scale real-time monitoring in the current detection methods for flavoring content and uniformity in tobacco, a characterization method for the uniformity of flavoring in tobacco based on hyperspectral detection had been developed. The method utilized a self-developed hyperspectral system to capture fluorescence hyperspectral images of tobacco before and after flavoring, and performed weighted unmixing on the fluorescence hyperspectral images. Based on the unmixing coefficients, a characterization index R for the content of flavoring applied to tobacco leaves was established, and the coefficient of variation (CV) of R was used to represent the uniformity of flavoring in the tobacco. Verification conducted on flavored tobacco leaves from actual production lines, it was found that the magnitude of R was positively correlated with the content of flavoring in the tobacco, and the CV could accurately determine the uniformity of flavoring in the tobacco. This method could be used for the detection and monitoring of flavoring quality of an known fragrance samples in the actual production process of tobacco.
To address the issues of complexity, low accuracy, high personnel requirements, and inability to adapt to the demand for large-scale real-time monitoring in the current detection methods for flavoring content and uniformity in tobacco, a characterization method for the uniformity of flavoring in tobacco based on hyperspectral detection had been developed. The method utilized a self-developed hyperspectral system to capture fluorescence hyperspectral images of tobacco before and after flavoring, and performed weighted unmixing on the fluorescence hyperspectral images. Based on the unmixing coefficients, a characterization index R for the content of flavoring applied to tobacco leaves was established, and the coefficient of variation (CV) of R was used to represent the uniformity of flavoring in the tobacco. Verification conducted on flavored tobacco leaves from actual production lines, it was found that the magnitude of R was positively correlated with the content of flavoring in the tobacco, and the CV could accurately determine the uniformity of flavoring in the tobacco. This method could be used for the detection and monitoring of flavoring quality of an known fragrance samples in the actual production process of tobacco.
2024, 39(5)
:102-108.
doi: 10.12187/2024.05.012
Abstract:
In order to study the availability of maltol-β-D-glucoside in cigarette flavoring, a comparative study was conducted on the stability, thermal pyrolytic behavior rate and fragrance application effect of maltol and maltol-β-D-glucoside. The results showed that the placement stability and thermal stability of maltol-β-D-glucoside were significantly higher than those of maltol. The main thermal cracking product of maltol-β-D-glucoside was maltol. Adding maltol-β-D-glycoside to cigarettes in different ways could achieve different flavoring effects. The thermal decomposition transfer rate of maltol-β-D-glucoside to mainstream smoke particulate matter was 1.46%, which was lower than the transfer rate of maltol (3.83%). However, cigarettes with added maltol-β-D-glucoside had better aroma release uniformity and stability than those with maltol. Therefore, maltol-β-D-glucoside was a precursor of flavors with good stability and aroma uniformity, which had certain advantages in the field of cigarette flavoring.
In order to study the availability of maltol-β-D-glucoside in cigarette flavoring, a comparative study was conducted on the stability, thermal pyrolytic behavior rate and fragrance application effect of maltol and maltol-β-D-glucoside. The results showed that the placement stability and thermal stability of maltol-β-D-glucoside were significantly higher than those of maltol. The main thermal cracking product of maltol-β-D-glucoside was maltol. Adding maltol-β-D-glycoside to cigarettes in different ways could achieve different flavoring effects. The thermal decomposition transfer rate of maltol-β-D-glucoside to mainstream smoke particulate matter was 1.46%, which was lower than the transfer rate of maltol (3.83%). However, cigarettes with added maltol-β-D-glucoside had better aroma release uniformity and stability than those with maltol. Therefore, maltol-β-D-glucoside was a precursor of flavors with good stability and aroma uniformity, which had certain advantages in the field of cigarette flavoring.
2024, 39(5)
:109-117.
doi: 10.12187/2024.05.013
Abstract:
To study the hygroscopicity of heated tobacco’s core materials, rolled tobacco sheet was taken as the research object, the effects of external environmental relative humidity, glycerol mass fraction and particle size of tobacco powder on its hygroscopicity were investigated through the single factor experiment. The differences in hygroscopicity of tobacco sheets prepared by rolling method, thick pulp method and paper making method were compared. The first-order and second-order adsorption kinetic models were used to fit and analyze the moisture absorption kinetic characteristics of rolled tobacco sheet. Further investigation was conducted on the hygroscopic thermodynamic properties of rolled tobacco sheet using six common adsorption models. The results showed that environmental relative humidity and mass fraction had the greatest impact on hygroscopicity, the higher the environmental relative humidity and the glycerol mass fraction, the stronger the hygroscopicity. The effect of particle size of tobacco powder on hygroscopicity was insignificant. The hygroscopicity of rolled tobacco sheet was comparable to that of tobacco slices produced by thick pulp method, and both were stronger than that of tobacco slices produced by paper making method. The moisture absorption time curve of rolled tobacco sheet was more in line with the second-order adsorption kinetics model, and its moisture absorption isotherm curve was more in line with the Peleg model.
To study the hygroscopicity of heated tobacco’s core materials, rolled tobacco sheet was taken as the research object, the effects of external environmental relative humidity, glycerol mass fraction and particle size of tobacco powder on its hygroscopicity were investigated through the single factor experiment. The differences in hygroscopicity of tobacco sheets prepared by rolling method, thick pulp method and paper making method were compared. The first-order and second-order adsorption kinetic models were used to fit and analyze the moisture absorption kinetic characteristics of rolled tobacco sheet. Further investigation was conducted on the hygroscopic thermodynamic properties of rolled tobacco sheet using six common adsorption models. The results showed that environmental relative humidity and mass fraction had the greatest impact on hygroscopicity, the higher the environmental relative humidity and the glycerol mass fraction, the stronger the hygroscopicity. The effect of particle size of tobacco powder on hygroscopicity was insignificant. The hygroscopicity of rolled tobacco sheet was comparable to that of tobacco slices produced by thick pulp method, and both were stronger than that of tobacco slices produced by paper making method. The moisture absorption time curve of rolled tobacco sheet was more in line with the second-order adsorption kinetics model, and its moisture absorption isotherm curve was more in line with the Peleg model.
2024, 39(5)
:118-126.
doi: 10.12187/2024.05.014
Abstract:
A U-shaped fixed bed reaction system was used to conduct pyrolysis experiments on 35 types of cigar tobaccos from eight different regions at home and abroad. The produced volatiles were analayzed in real-time by photoionization mass spectrometry (PI-MS). The principal component analysis (PCA) method was used to statistically analyze the obtained MS datas. The results showed that a total of 74 shared pyrolysis products were identified in the 35 types of cigar tobaccos, mainly hydrocarbons, phenols, ketones, alcohols, esters, and nitrogen-containing substances. There were significant differences in the composition of pyrolysis products between domestic and foreign cigar tobaccos. Especially, the domestic cigar coats had a much higher content of nicotine than the foreign cigar coats, whereas the relative content of fragrant components such as methylcyclopentenone, 2,3-dimethyl-2-cyclopentenone, and 2,5-dimethylcyclopentenone in the pyrolysis products of domestic cigar coats was lower than that of foreign cigar coats. The relative content of fragrant components such as 3-methylpyridine, 2,5-dimethylcyclopentenone, and (+)-limonene in the pyrolysis products of domestic cigar chips was lower than that of foreign cigar chips. In addition, the characteristic pyrolysis products of cigar tobaccos screened by PCA were consistent with the results of routine chemical composition analysis of cigar tobaccos, confirming the accuracy of the experimental analysis results.
A U-shaped fixed bed reaction system was used to conduct pyrolysis experiments on 35 types of cigar tobaccos from eight different regions at home and abroad. The produced volatiles were analayzed in real-time by photoionization mass spectrometry (PI-MS). The principal component analysis (PCA) method was used to statistically analyze the obtained MS datas. The results showed that a total of 74 shared pyrolysis products were identified in the 35 types of cigar tobaccos, mainly hydrocarbons, phenols, ketones, alcohols, esters, and nitrogen-containing substances. There were significant differences in the composition of pyrolysis products between domestic and foreign cigar tobaccos. Especially, the domestic cigar coats had a much higher content of nicotine than the foreign cigar coats, whereas the relative content of fragrant components such as methylcyclopentenone, 2,3-dimethyl-2-cyclopentenone, and 2,5-dimethylcyclopentenone in the pyrolysis products of domestic cigar coats was lower than that of foreign cigar coats. The relative content of fragrant components such as 3-methylpyridine, 2,5-dimethylcyclopentenone, and (+)-limonene in the pyrolysis products of domestic cigar chips was lower than that of foreign cigar chips. In addition, the characteristic pyrolysis products of cigar tobaccos screened by PCA were consistent with the results of routine chemical composition analysis of cigar tobaccos, confirming the accuracy of the experimental analysis results.
Journal Information
Founded in 1986， bimonthly
Administered by：The Education Department Henan Province
Sponsored by：Zhengzhou University of Light Industry
Editor-in-chief：Wei Shizhong
Executive Editor-in-Chief：Zou Lin
Deputy Editor-in-Chief：Qu Shuanghong
Edited & published by：Editorial Department of Journal of Light Industry
CN 41-1437/TS
ISSN 2096-1553
Address：136 Science Avenue, Zhengzhou City, Henan Province, China
Postal Code：450001
Tel：(086)0371-86608635
(086)0371-86608633
