Pub Date : 2024-01-10DOI: 10.1177/09670335231225820
Rahul Sreekumar, N. Ashwath, D. Cozzolino, KB Walsh
This study was conducted to evaluate the ability of near infrared (NIR) spectroscopy to estimate oil content, and per cent of cake, resin and residue in beauty leaf tree ( Calophyllum inophyllum L.) kernel samples. Fruits were collected from various geographical locations of tropical Australia (from Rockhampton to Darwin) and air dried before the kernels were manually separated from the fruits. Kernel samples were oven dried, crushed (5–10 mm) and their NIR spectra collected using a Fourier transform (FT) NIR instrument where the same batch of kernels were used to extract oil using a screw press. Calibration models between the NIR spectra and reference data were developed using partial least squares (PLS) regression. The cross-validation statistics including the coefficient of determination (r2) and standard error in cross validation (SECV) were 0.83 (SECV: 2.39%) for oil content, 0.89 (SECV: 2.81%) for cake, 0.88 (SECV: 1.92%) for resin and 0.79 (SECV: 2.15%) for residue, respectively. This research showed that NIR spectroscopy can be used as an alternative, faster and low-cost technique to predict oil content, per cent of cake, resins and residues in various genotypes of beauty leaf tree. Further studies should be carried out to increase the sample size and chemical variation, as well as to evaluate different methods of oil extraction (e.g., solvent extraction) to improve the reliability of the calibration models.
{"title":"Predicting oil content of Australian beauty leaf tree kernel samples using near infrared spectroscopy combined with chemometrics","authors":"Rahul Sreekumar, N. Ashwath, D. Cozzolino, KB Walsh","doi":"10.1177/09670335231225820","DOIUrl":"https://doi.org/10.1177/09670335231225820","url":null,"abstract":"This study was conducted to evaluate the ability of near infrared (NIR) spectroscopy to estimate oil content, and per cent of cake, resin and residue in beauty leaf tree ( Calophyllum inophyllum L.) kernel samples. Fruits were collected from various geographical locations of tropical Australia (from Rockhampton to Darwin) and air dried before the kernels were manually separated from the fruits. Kernel samples were oven dried, crushed (5–10 mm) and their NIR spectra collected using a Fourier transform (FT) NIR instrument where the same batch of kernels were used to extract oil using a screw press. Calibration models between the NIR spectra and reference data were developed using partial least squares (PLS) regression. The cross-validation statistics including the coefficient of determination (r2) and standard error in cross validation (SECV) were 0.83 (SECV: 2.39%) for oil content, 0.89 (SECV: 2.81%) for cake, 0.88 (SECV: 1.92%) for resin and 0.79 (SECV: 2.15%) for residue, respectively. This research showed that NIR spectroscopy can be used as an alternative, faster and low-cost technique to predict oil content, per cent of cake, resins and residues in various genotypes of beauty leaf tree. Further studies should be carried out to increase the sample size and chemical variation, as well as to evaluate different methods of oil extraction (e.g., solvent extraction) to improve the reliability of the calibration models.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"64 18","pages":""},"PeriodicalIF":1.8,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139441164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-09DOI: 10.1177/09670335231213951
Rumbidzai T Matenda, Diane Rip, Paul J Williams
Near infrared (NIR) hyperspectral imaging and multivariate data analysis was evaluated for its potential to detect and classify Listeria species. Three Listeria species, namely L. monocytogenes (ATCC 23074), L. innocua (ATCC 33090) and L. ivanovii (ATCC 19119) were grown for single colonies on Brain Heart Infusion agar and imaged in the NIR range of 950–2500 nm. Principal component analysis (PCA) was used for data exploration and to establish pattern recognition. Images were pre-processed with standard normal variate correction and the Savitzky-Golay smoothing technique (third order polynomial with 15 points). Two approaches to data analysis, that is object-wise and pixel-wise analysis, were investigated for discriminant analysis. The PCA score plot showed slight separation between the three groups with L. monocytogenes and L. ivanovii grouping close together. It was possible to visualise separation along PC3 (5.64% sum of squares (SS)) and PC4 (3.44% SS). Based on the loadings, differences in bacteria were attributed to teichoic acids, protein, and carbohydrate composition in the bacterial cell wall within the wavelength range 1000–1900 nm. Using extracted spectral data from the hypercubes, partial least squares discriminant analysis was employed for further classification. Classification accuracies above 90% were achieved for L. monocytogenes, L. innocua and L. ivanovii. This was true for data analysed using both pixel-wise analysis and object-wise analysis. The results demonstrated that hyperspectral imaging has notable potential to classify bacteria within the Listeria genus. Nonetheless, in order to improve model efficiency, model optimisation and incorporation of more bacterial strains need to be investigated in further research.
{"title":"Classification of <i>Listeria</i> species using near infrared hyperspectral imaging","authors":"Rumbidzai T Matenda, Diane Rip, Paul J Williams","doi":"10.1177/09670335231213951","DOIUrl":"https://doi.org/10.1177/09670335231213951","url":null,"abstract":"Near infrared (NIR) hyperspectral imaging and multivariate data analysis was evaluated for its potential to detect and classify Listeria species. Three Listeria species, namely L. monocytogenes (ATCC 23074), L. innocua (ATCC 33090) and L. ivanovii (ATCC 19119) were grown for single colonies on Brain Heart Infusion agar and imaged in the NIR range of 950–2500 nm. Principal component analysis (PCA) was used for data exploration and to establish pattern recognition. Images were pre-processed with standard normal variate correction and the Savitzky-Golay smoothing technique (third order polynomial with 15 points). Two approaches to data analysis, that is object-wise and pixel-wise analysis, were investigated for discriminant analysis. The PCA score plot showed slight separation between the three groups with L. monocytogenes and L. ivanovii grouping close together. It was possible to visualise separation along PC3 (5.64% sum of squares (SS)) and PC4 (3.44% SS). Based on the loadings, differences in bacteria were attributed to teichoic acids, protein, and carbohydrate composition in the bacterial cell wall within the wavelength range 1000–1900 nm. Using extracted spectral data from the hypercubes, partial least squares discriminant analysis was employed for further classification. Classification accuracies above 90% were achieved for L. monocytogenes, L. innocua and L. ivanovii. This was true for data analysed using both pixel-wise analysis and object-wise analysis. The results demonstrated that hyperspectral imaging has notable potential to classify bacteria within the Listeria genus. Nonetheless, in order to improve model efficiency, model optimisation and incorporation of more bacterial strains need to be investigated in further research.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":" 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135241250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-08DOI: 10.1177/09670335231211618
Zhonghai He, Kun Shen, Xiaofang Zhang
In multivariate calibration problems, model performance is affected significantly by the calibration samples used during model building. In recent years, active learning methods have become one of the best methods for sample selection. However, most active learning methods only select instances from prediction uncertainty or sample space distance, and these single-criteria methods tend to select undesired samples. In addition, sample density characterizes the spatial information carried by the sample, but few studies in quantitative analysis utilize sample density alone to select calibration samples. Considering these issues, based on the k-means clustering algorithm, this paper proposes an active learning sample selection method (DIDAL), which combines the three criteria of diversity, informativeness and sample density. The most representative sample is iteratively selected for - addition to the calibration set for modeling and estimating the chemical concentration of analytes. Soybean meal and soy sauce samples were analyzed by DIDAL and compared with existing sample selection methods. The prediction results show that the DIDAL algorithm significantly outperforms several existing algorithms and is close to the performance of full-sample modeling. A model with high prediction accuracy can be constructed by selecting only a few samples using the DIDAL method.
{"title":"Active learning sample selection - based on multicriteria","authors":"Zhonghai He, Kun Shen, Xiaofang Zhang","doi":"10.1177/09670335231211618","DOIUrl":"https://doi.org/10.1177/09670335231211618","url":null,"abstract":"In multivariate calibration problems, model performance is affected significantly by the calibration samples used during model building. In recent years, active learning methods have become one of the best methods for sample selection. However, most active learning methods only select instances from prediction uncertainty or sample space distance, and these single-criteria methods tend to select undesired samples. In addition, sample density characterizes the spatial information carried by the sample, but few studies in quantitative analysis utilize sample density alone to select calibration samples. Considering these issues, based on the k-means clustering algorithm, this paper proposes an active learning sample selection method (DIDAL), which combines the three criteria of diversity, informativeness and sample density. The most representative sample is iteratively selected for - addition to the calibration set for modeling and estimating the chemical concentration of analytes. Soybean meal and soy sauce samples were analyzed by DIDAL and compared with existing sample selection methods. The prediction results show that the DIDAL algorithm significantly outperforms several existing algorithms and is close to the performance of full-sample modeling. A model with high prediction accuracy can be constructed by selecting only a few samples using the DIDAL method.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"144 3‐6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135392765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.1177/09670335231200998
Alexandre E Santos, Laiz R Ventura, Carlos E Fellows
A new study of the 2−0 band of the hydrochloric acid molecule is performed by high resolution Fourier-transform absorption spectroscopy in the near infrared region. The spectra were measured at two different temperatures, 293 K and 315 K, for different pressures at each temperature. The spectral linewidths were analysed in a two-step procedure, being first performed by directly measuring the linewidth and second by fitting each spectral line to a model line profile, using Gaussian, Loretzian and Voigt profiles. A study of the profiles that best describe the spectral line fits is carried out in this work. The behavior of the spectral lines self-broadening and their corresponding self-induced shifts were studied for different values of rotational quantum numbers. The analysis are performed for both isotopes of the molecule and the self-broadening and self-shift coefficients are presented.
{"title":"A new hydrochloric acid 2-0 band analysis: A two temperature study","authors":"Alexandre E Santos, Laiz R Ventura, Carlos E Fellows","doi":"10.1177/09670335231200998","DOIUrl":"https://doi.org/10.1177/09670335231200998","url":null,"abstract":"A new study of the 2−0 band of the hydrochloric acid molecule is performed by high resolution Fourier-transform absorption spectroscopy in the near infrared region. The spectra were measured at two different temperatures, 293 K and 315 K, for different pressures at each temperature. The spectral linewidths were analysed in a two-step procedure, being first performed by directly measuring the linewidth and second by fitting each spectral line to a model line profile, using Gaussian, Loretzian and Voigt profiles. A study of the profiles that best describe the spectral line fits is carried out in this work. The behavior of the spectral lines self-broadening and their corresponding self-induced shifts were studied for different values of rotational quantum numbers. The analysis are performed for both isotopes of the molecule and the self-broadening and self-shift coefficients are presented.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"202 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136213034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-26DOI: 10.1177/09670335231202258
Mohammad Nadimi, Fernando AM Saccon, Ahmed Elrewainy, Dennis Parcey, Sherif S Sherif, Jitendra Paliwal
With the continuously growing world population in the 21st century, the agri-food industry is in dire need of adopting rapid, eco-friendly, and reliable technologies to improve the quantity, quality, and safety of agri-food products to fulfill the world's future food needs. Hyperspectral imaging (HSI), a technique to glean a sample's spectral and spatial information, is an emerging non-destructive technique that can characterize the quality parameters of agri-food products such as Fusarium damage. Despite its vast potential, HSI systems suffer from enormous data sizes, requiring high computational time and power. One potential solution to overcome the aforementioned challenge is to reduce the data size by removing redundant information. However, detecting small optimum features from a large dataset is not trivial. To this end, an exploratory novel HSI data reduction and analysis technique was investigated and validated to identify Fusarium damage in wheat kernels. Wheat samples at three moisture contents (19, 27, and 35%, wet basis) and seven infection levels (ranging from 0 to 56 days after infection) were imaged at 256 equally spaced wavelengths from 820 to 1666 nm. Firstly, complete HSI data was utilized to successfully characterize sound and Fusarium-damaged wheat kernels using independent component analysis (ICA) algorithm. Then, a genetic algorithm optimization approach was used to reduce the data to ten wavelengths for ICA-based analysis. This data reduction approach reduced the computation time to approximately 1.31% of the original time taken for analyzing the full HSI data without compromising the performance of the system. This preliminary study suggests that such wavelength tailoring could reduce the complexity and price of the imaging hardware, e.g., the use of inexpensive non-tunable filters, and less expensive computational hardware, thereby enabling fast and affordable real-time exploration and sorting of grains. This study, while exploratory, fosters advancements in HSI data processing and identifies certain limitations that open new avenues for future research.
{"title":"Investigation of <i>Fusarium</i> damage in wheat using hyperspectral imaging: An independent component analysis approach","authors":"Mohammad Nadimi, Fernando AM Saccon, Ahmed Elrewainy, Dennis Parcey, Sherif S Sherif, Jitendra Paliwal","doi":"10.1177/09670335231202258","DOIUrl":"https://doi.org/10.1177/09670335231202258","url":null,"abstract":"With the continuously growing world population in the 21st century, the agri-food industry is in dire need of adopting rapid, eco-friendly, and reliable technologies to improve the quantity, quality, and safety of agri-food products to fulfill the world's future food needs. Hyperspectral imaging (HSI), a technique to glean a sample's spectral and spatial information, is an emerging non-destructive technique that can characterize the quality parameters of agri-food products such as Fusarium damage. Despite its vast potential, HSI systems suffer from enormous data sizes, requiring high computational time and power. One potential solution to overcome the aforementioned challenge is to reduce the data size by removing redundant information. However, detecting small optimum features from a large dataset is not trivial. To this end, an exploratory novel HSI data reduction and analysis technique was investigated and validated to identify Fusarium damage in wheat kernels. Wheat samples at three moisture contents (19, 27, and 35%, wet basis) and seven infection levels (ranging from 0 to 56 days after infection) were imaged at 256 equally spaced wavelengths from 820 to 1666 nm. Firstly, complete HSI data was utilized to successfully characterize sound and Fusarium-damaged wheat kernels using independent component analysis (ICA) algorithm. Then, a genetic algorithm optimization approach was used to reduce the data to ten wavelengths for ICA-based analysis. This data reduction approach reduced the computation time to approximately 1.31% of the original time taken for analyzing the full HSI data without compromising the performance of the system. This preliminary study suggests that such wavelength tailoring could reduce the complexity and price of the imaging hardware, e.g., the use of inexpensive non-tunable filters, and less expensive computational hardware, thereby enabling fast and affordable real-time exploration and sorting of grains. This study, while exploratory, fosters advancements in HSI data processing and identifies certain limitations that open new avenues for future research.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134960507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-25DOI: 10.1177/09670335231194739
Babirye Fatumah Namakula, Ephraim Nuwamanya, Michael Kanaabi, Enoch Wembambazi, Robert Sezi Kawuki
In Uganda, efforts are underway to improve starch content through conventional breeding as a strategy for increasing adoption of new cassava varieties for both food and industry. However, only few samples can be quantified, limiting the gains in breeding. A database of 115 clones was used to evaluate the potential of Near infrared spectroscopy to predict starch content in cassava. Starch content ranged from 21.48 to 73.97% dry basis. The performance of standard normal variate and de-trend with second derivative calculated on two data points and smoothing plus combination of standard multiplicative scatter correction with second derivative calculated on two data points and smoothing were the best fit mathematical treatments for the calibrations developed. Near infrared spectroscopy predictions for starch content (R 2 = 0.85, and r 2 = 0.55) developed were reliable, thus usable for screening of cassava starch content at early stages of breeding.
{"title":"Predicting starch content of cassava with near infrared spectroscopy in Ugandan cassava germplasm","authors":"Babirye Fatumah Namakula, Ephraim Nuwamanya, Michael Kanaabi, Enoch Wembambazi, Robert Sezi Kawuki","doi":"10.1177/09670335231194739","DOIUrl":"https://doi.org/10.1177/09670335231194739","url":null,"abstract":"In Uganda, efforts are underway to improve starch content through conventional breeding as a strategy for increasing adoption of new cassava varieties for both food and industry. However, only few samples can be quantified, limiting the gains in breeding. A database of 115 clones was used to evaluate the potential of Near infrared spectroscopy to predict starch content in cassava. Starch content ranged from 21.48 to 73.97% dry basis. The performance of standard normal variate and de-trend with second derivative calculated on two data points and smoothing plus combination of standard multiplicative scatter correction with second derivative calculated on two data points and smoothing were the best fit mathematical treatments for the calibrations developed. Near infrared spectroscopy predictions for starch content (R 2 = 0.85, and r 2 = 0.55) developed were reliable, thus usable for screening of cassava starch content at early stages of breeding.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135864717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.1177/09670335231193117
N. Pretorius, Ashley T. Forrest, K. Walsh
Hydroxyapatite is a major component of teeth and bones and is used commercially in metal sequestration. Near infrared imaging and spectroscopy has found increasing use in characterisation of these materials, particularly in context of dental conditions. The near infrared spectra of these materials are reviewed in terms of band assignments related to water in various states, P-OH and organic material, and in terms of light scattering. The effect of factors such as acid and heat on the NIR spectra of bones and teeth is also described. This review is intended to provide a resource for future researchers using NIR spectroscopy in characterisation of hydroxyapatite containing material.
{"title":"A review of near infrared spectroscopic features of teeth, bone and artificial hydroxyapatite","authors":"N. Pretorius, Ashley T. Forrest, K. Walsh","doi":"10.1177/09670335231193117","DOIUrl":"https://doi.org/10.1177/09670335231193117","url":null,"abstract":"Hydroxyapatite is a major component of teeth and bones and is used commercially in metal sequestration. Near infrared imaging and spectroscopy has found increasing use in characterisation of these materials, particularly in context of dental conditions. The near infrared spectra of these materials are reviewed in terms of band assignments related to water in various states, P-OH and organic material, and in terms of light scattering. The effect of factors such as acid and heat on the NIR spectra of bones and teeth is also described. This review is intended to provide a resource for future researchers using NIR spectroscopy in characterisation of hydroxyapatite containing material.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"31 1","pages":"227 - 240"},"PeriodicalIF":1.8,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43844075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pericarpium Citri Reticulatae is a traditional Chinese medicine with high medicinal value, and its storage age has a great impact on its ethno-pharmaceutical relevance. At present, there is a situation in the market place where Pericarpium Citri Reticulatae with short storage age is fraudulently sold as Pericarpium Citri Reticulatae with long storage age, and some unaged orange peels dyed with tea are sold as Pericarpium Citri Reticulatae at a high price. In this study, a rapid, on-site method for identifying the storage age of Xinhui Pericarpium Citri Reticulatae based on spectral imaging technology was described. The image features and spectral features were extracted respectively from the surface reflection spectral images of Pericarpium Citri Reticulatae, and a machine learning model was established to identify the storage age. This study explored the classification effect of the combination of different spectral pre-processing methods and machine learning models, and finally selected the combination of standard normal variate and random forest models, to achieve 95% accuracy on the test dataset, showing excellent generalization performance. The result shows that the spectral imaging technology can rapidly identify the storage age of Xinhui Pericarpium Citri Reticulatae in real time, which has a great application prospect in the detection of the properties of medicinal materials.
{"title":"On-site rapid detection of aging of Pericarpium Citri Reticulatae using multispectral imaging","authors":"Yuchen Guo, Xiangyang Yu, Weibin Hong, Yefan Cai, Wanbang Xu, hongyu Gu","doi":"10.1177/09670335231194737","DOIUrl":"https://doi.org/10.1177/09670335231194737","url":null,"abstract":"Pericarpium Citri Reticulatae is a traditional Chinese medicine with high medicinal value, and its storage age has a great impact on its ethno-pharmaceutical relevance. At present, there is a situation in the market place where Pericarpium Citri Reticulatae with short storage age is fraudulently sold as Pericarpium Citri Reticulatae with long storage age, and some unaged orange peels dyed with tea are sold as Pericarpium Citri Reticulatae at a high price. In this study, a rapid, on-site method for identifying the storage age of Xinhui Pericarpium Citri Reticulatae based on spectral imaging technology was described. The image features and spectral features were extracted respectively from the surface reflection spectral images of Pericarpium Citri Reticulatae, and a machine learning model was established to identify the storage age. This study explored the classification effect of the combination of different spectral pre-processing methods and machine learning models, and finally selected the combination of standard normal variate and random forest models, to achieve 95% accuracy on the test dataset, showing excellent generalization performance. The result shows that the spectral imaging technology can rapidly identify the storage age of Xinhui Pericarpium Citri Reticulatae in real time, which has a great application prospect in the detection of the properties of medicinal materials.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"31 1","pages":"263 - 270"},"PeriodicalIF":1.8,"publicationDate":"2023-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43510950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-02DOI: 10.1177/09670335231193113
H. Zayani, Y. Fouad, D. Michot, Z. Kassouk, Z. Lili-Chabaane, C. Walter
Monitoring changes in soil properties is essential to ensure ecosystem function and agricultural productivity. This study evaluated the ability of visible near infrared (Vis-NIR) spectroscopy to detect the temporal trend in soil organic carbon (SOC) content after 5 years in a 12 km2 agricultural catchment in western France. Partial least squares regression models were developed using soil samples from a local dataset collected in 2013 at two depths (198 samples at 0–15 cm and 196 samples at 15–25 cm) to predict SOC content of 111 new samples collected in 2018 at the same locations and at similar depths (0–15 cm and 15–25 cm). Two approaches, which differed in whether or not they considered the SOC content variability that can result from collecting soil samples at two depths, were applied. For both approaches, the potential benefit of “temporal spiking” was evaluated by adding 10% of 2018 samples to the 2013 dataset. The results showed that removing outliers and stratifying the calibration dataset by depth yielded the highest accuracy, with SOC RMSEP of 4.1 and 2.7 g.kg−1 for 0–15 and 15–25 cm, respectively. Moreover, temporal spiking improved five of eight predictions (stratifying or not the calibration dataset by depth, removing or not poorly predicted outliers), with increases in the ratio of performance to deviation of 0.10–0.44. Furthermore, comparing observed and predicted changes in SOC content showed that Vis-NIR spectroscopy estimated its trend over time in most cases.
{"title":"Detecting the temporal trend of cultivated soil organic carbon content using visible near infrared spectroscopy","authors":"H. Zayani, Y. Fouad, D. Michot, Z. Kassouk, Z. Lili-Chabaane, C. Walter","doi":"10.1177/09670335231193113","DOIUrl":"https://doi.org/10.1177/09670335231193113","url":null,"abstract":"Monitoring changes in soil properties is essential to ensure ecosystem function and agricultural productivity. This study evaluated the ability of visible near infrared (Vis-NIR) spectroscopy to detect the temporal trend in soil organic carbon (SOC) content after 5 years in a 12 km2 agricultural catchment in western France. Partial least squares regression models were developed using soil samples from a local dataset collected in 2013 at two depths (198 samples at 0–15 cm and 196 samples at 15–25 cm) to predict SOC content of 111 new samples collected in 2018 at the same locations and at similar depths (0–15 cm and 15–25 cm). Two approaches, which differed in whether or not they considered the SOC content variability that can result from collecting soil samples at two depths, were applied. For both approaches, the potential benefit of “temporal spiking” was evaluated by adding 10% of 2018 samples to the 2013 dataset. The results showed that removing outliers and stratifying the calibration dataset by depth yielded the highest accuracy, with SOC RMSEP of 4.1 and 2.7 g.kg−1 for 0–15 and 15–25 cm, respectively. Moreover, temporal spiking improved five of eight predictions (stratifying or not the calibration dataset by depth, removing or not poorly predicted outliers), with increases in the ratio of performance to deviation of 0.10–0.44. Furthermore, comparing observed and predicted changes in SOC content showed that Vis-NIR spectroscopy estimated its trend over time in most cases.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"31 1","pages":"241 - 255"},"PeriodicalIF":1.8,"publicationDate":"2023-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48883941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-19DOI: 10.1177/09670335231168437
L. Ni, Zhange Zhang, Liguo Zhang, S. Luan
Two case studies were conducted to verify calibration model transfer methods without standards by multi-step wavelength selection, using 3–7 near infrared spectrometers to predict ingredients in corn and total plant alkaloids (TPA) in tobacco leaves. Based on the characteristic wavelengths of Uc, which are selected using the scale-invariant feature transform (SIFT), this study advances two multistep wavelength selection methods by selecting wavelengths with high independence and a high standard deviation of the sample spectra (SDSS). The first method, SIFT-SDSS-CORX, selects important characteristic wavelengths Uc-i from Uc whose SDSS is greater than a threshold SDSSacrit. Subsequently, rx, the correlation coefficient matrix between spectral signals of Uc-i, is calculated, and only one wavelength is retained from those whose correlation coefficients exceed a threshold, rxacrit. The wavelength set Uc-i-rx, which is finally screened, is important and independent. In the second method, SIFT-CORX-SDSS, Uc-rx is first selected from Uc by retaining only one wavelength from those whose correlation coefficients between spectral signals of Uc exceed a threshold, rxbcrit. Subsequently, the wavelengths Uc-rx-i with SDSS exceeding a threshold SDSSbcrit are selected from Uc-rx. Near infrared spectroscopy calibration models for predicting protein and oil in corn and TPA in tobacco leaves were built using partial least squares regression (PLS) based on different wavelength sets of Uc, Uc-i, Uc-i-rx, Uc-rx, and Uc-rx-i, respectively. The latent variables used in the PLS models were determined by an accumulative contribution ratio over 99.9%. The results indicate that the PLS models built on Uc-i-rx and Uc-rx-i are effective on both primary and secondary units for corn and tobacco samples. This study utilises a three-step wavelength selection method to select highly independent, important, and characteristic spectral variables, thereby enhancing the robustness, simplicity, and interpretability of NIR) calibration models and facilitating their transfer to secondary units without standards.
{"title":"Transferring near infrared spectral calibration models without standards via multistep wavelength selection","authors":"L. Ni, Zhange Zhang, Liguo Zhang, S. Luan","doi":"10.1177/09670335231168437","DOIUrl":"https://doi.org/10.1177/09670335231168437","url":null,"abstract":"Two case studies were conducted to verify calibration model transfer methods without standards by multi-step wavelength selection, using 3–7 near infrared spectrometers to predict ingredients in corn and total plant alkaloids (TPA) in tobacco leaves. Based on the characteristic wavelengths of Uc, which are selected using the scale-invariant feature transform (SIFT), this study advances two multistep wavelength selection methods by selecting wavelengths with high independence and a high standard deviation of the sample spectra (SDSS). The first method, SIFT-SDSS-CORX, selects important characteristic wavelengths Uc-i from Uc whose SDSS is greater than a threshold SDSSacrit. Subsequently, rx, the correlation coefficient matrix between spectral signals of Uc-i, is calculated, and only one wavelength is retained from those whose correlation coefficients exceed a threshold, rxacrit. The wavelength set Uc-i-rx, which is finally screened, is important and independent. In the second method, SIFT-CORX-SDSS, Uc-rx is first selected from Uc by retaining only one wavelength from those whose correlation coefficients between spectral signals of Uc exceed a threshold, rxbcrit. Subsequently, the wavelengths Uc-rx-i with SDSS exceeding a threshold SDSSbcrit are selected from Uc-rx. Near infrared spectroscopy calibration models for predicting protein and oil in corn and TPA in tobacco leaves were built using partial least squares regression (PLS) based on different wavelength sets of Uc, Uc-i, Uc-i-rx, Uc-rx, and Uc-rx-i, respectively. The latent variables used in the PLS models were determined by an accumulative contribution ratio over 99.9%. The results indicate that the PLS models built on Uc-i-rx and Uc-rx-i are effective on both primary and secondary units for corn and tobacco samples. This study utilises a three-step wavelength selection method to select highly independent, important, and characteristic spectral variables, thereby enhancing the robustness, simplicity, and interpretability of NIR) calibration models and facilitating their transfer to secondary units without standards.","PeriodicalId":16551,"journal":{"name":"Journal of Near Infrared Spectroscopy","volume":"31 1","pages":"171 - 185"},"PeriodicalIF":1.8,"publicationDate":"2023-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45479809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：