Determination of the total volatile basic nitrogen (TVB-N) content in pork meat using hyperspectral fluorescence imaging

Lee, Hoonsoo; Kim, Moon S.; Lee, Wang‐Hee; Cho, Byoung‐Kwan

doi:10.1016/j.snb.2017.12.102

Cited by 80 publications

(30 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Meat with TVB‐N value ≤15 mg/100 g is considered fresh, 15–20 mg/100 g is cooking edible but less fresh, and ≥20 mg/100 g indicates inedible and spoiled. [ 41–43 ] Then these images were classified according to these three standard categories of freshness—fresh, less fresh, and spoiled. As there are variations of TVB‐N threshold for meat with different types and sources, a detailed investigation of TVB‐N values of a specific meat is suggested to ensure the accurate classification.…”

Section: Figurementioning

confidence: 99%

Portable Food‐Freshness Prediction Platform Based on Colorimetric Barcode Combinatorics and Deep Convolutional Neural Networks

et al. 2020

View full text Add to dashboard Cite

Artificial scent screening systems (known as electronic noses, E‐noses) have been researched extensively. A portable, automatic, and accurate, real‐time E‐nose requires both robust cross‐reactive sensing and fingerprint pattern recognition. Few E‐noses have been commercialized because they suffer from either sensing or pattern‐recognition issues. Here, cross‐reactive colorimetric barcode combinatorics and deep convolutional neural networks (DCNNs) are combined to form a system for monitoring meat freshness that concurrently provides scent fingerprint and fingerprint recognition. The barcodes—comprising 20 different types of porous nanocomposites of chitosan, dye, and cellulose acetate—form scent fingerprints that are identifiable by DCNN. A fully supervised DCNN trained using 3475 labeled barcode images predicts meat freshness with an overall accuracy of 98.5%. Incorporating DCNN into a smartphone application forms a simple platform for rapid barcode scanning and identification of food freshness in real time. The system is fast, accurate, and non‐destructive, enabling consumers and all stakeholders in the food supply chain to monitor food freshness.

show abstract

Section: Figurementioning

confidence: 99%

Portable Food‐Freshness Prediction Platform Based on Colorimetric Barcode Combinatorics and Deep Convolutional Neural Networks

et al. 2020

View full text Add to dashboard Cite

show abstract

“…can be tended to the nonlinearity. Similarly, several previous studies reported that the superior of nonlinear regression models over linear models for predicting various food quality and safety indicators (Cheng & Sun, 2015b;Cheng, Sun, Pu, & Zhu, 2015;Cheng et al, 2017;Lee, Kim, Lee, & Cho, 2018).…”

Section: Thus Relationship Between Tbars Values and The Hypercube Datamentioning

confidence: 76%

Comparison of various chemometric analysis for rapid prediction of thiobarbituric acid reactive substances in rainbow trout fillets by hyperspectral imaging technique

Khoshnoudi‐Nia

Moosavi‐Nasab

2019

Food Science & Nutrition

View full text Add to dashboard Cite

This study explores the potential application of hyperspectral imaging (HSI; 430–1,010 nm) coupled with different linear and nonlinear models for rapid nondestructive evaluation of thiobarbituric acid‐reactive substances (TBARS) value in rainbow trout ( Oncorhynchus mykiss) fillets during 12 days of cold storage (4 ± 2°C). HSI data and TBARS value of fillets were obtained in the laboratory. The primary prediction models were established based on linear partial least squares regression (PLSR) and least squares support vector machine (LS‐SVM). In full spectral range, the prediction capability of LS‐SVM ( = 0.829; RMSEP = 0.128 mg malondialdehyde [MDA]/kg) was better than PLSR ( = 0.748; RMSEP = 0.155 mg MDA/kg) model and LS‐SVM model exhibited satisfactory prediction performance ( > 0.82). To simplify the calibration models, a combination of uninformative variable elimination and backward regression (UB) was used as variable selection. Nine wavelengths were selected. Various chemometric analysis methods including linear PLSR and multiple linear regression and nonlinear LS‐SVM and back‐propagation artificial neural network (BP‐ANN) were compared. The simplified models showed better capability than those were built based on the whole dataset in prediction of TBARS values. Moreover, the nonlinear models were preferred over linear models. Among the four chemometric algorithms, the best and weakest models were LS‐SVM and PLSR model, respectively. UB‐LS‐SVM model was the optimal models for predicting TBARS value in rainbow trout fillets ( = 0.831; RMSEP = 0.125 mg MDA/kg). The establishing of lipid‐oxidation prediction model in rainbow trout fish was complicated, due to the fluctuations of TBARS values during storage. Therefore, further researches are needed to improve the prediction results and applicability of HIS technique for prediction of TBARS value in rainbow trout fish.

show abstract

“…In this study, the size of one ROI image was about 8,500 pixels, and each pixel spectrum included 1,288 wavelengths. Considering the redundancy and collinearity of hyperspectral data, it is essential to reduce the dimension of spectral data (Lee, Kim, Lee, & Cho, ; Siripatrawan & Makino, ). Several wavelengths (variables) selection methods, namely, uninformative variable elimination (UVE), competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), regression coefficients (RC), and the combination of different variable algorithms, were used to choose characteristic wavelengths from full spectrum ranges for optimizing the modeling procedures.…”

Section: Methodsmentioning

confidence: 99%

Total phenolic content prediction inFlos Loniceraeusing hyperspectral imaging combined with wavelengths selection methods

Liu

Wang

Gao

et al. 2019

J Food Process Engineering

View full text Add to dashboard Cite

Hyperspectral imaging (HSI) method was applied to rapidly and nondestructively predict total phenolic content in Flos Lonicerae. The least squares support vector machine (LS‐SVM) and partial least squares regression (PLSR) models were developed on the basis of full wavelengths data and characteristic wavelengths data chosen by six wavelengths selection ways. The results clarified that standard normal variable (SNV) was the optimal pretreatment method, and the nonlinear LS‐SVM model based on the characteristic wavelengths chosen by the combination (CARS‐SPA) of competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) yielded the best prediction performance for the total phenolic content. The overall results demonstrated that the proposed CARS‐SPA was a good method for selecting characteristic wavelengths to enhance prediction performance of HSI, and nonlinear LS‐SVM was more appropriate than linear PLSR for the prediction of total phenolic content in Flos Lonicerae. Practical applications Flos Lonicerae is a well‐known traditional Chinese medicinal herb, and also the raw material of various medicines and herbal tea. The demand for Flos Lonicerae is increasing. Therefore, monitoring the quality of Flos Lonicerae is very important to consumers and industry of Flos Lonicerae. The phenolic compounds content in Flos Lonicerae is one of the key internal quality factors. The traditional methods (such as spectrophotometric techniques or high‐performance liquid chromatography) for detecting total phenolic content are time‐consuming, labor‐consuming, and destructive. Hyperspectral imaging (HSI) is a nondestructive and reliable method for quality inspection. The results showed that SNV was determined as the best method for pretreatment, and the nonlinear CARS‐SPA‐LS‐SVM model was developed for total phenolic content detection with high precision. It could be verified that HSI technology is effective and promising for rapid determination of total phenolic content and quality control of Flos Lonicerae as well as other agricultural products.

show abstract

Determination of the total volatile basic nitrogen (TVB-N) content in pork meat using hyperspectral fluorescence imaging

Cited by 80 publications

References 53 publications

Portable Food‐Freshness Prediction Platform Based on Colorimetric Barcode Combinatorics and Deep Convolutional Neural Networks

Portable Food‐Freshness Prediction Platform Based on Colorimetric Barcode Combinatorics and Deep Convolutional Neural Networks

Comparison of various chemometric analysis for rapid prediction of thiobarbituric acid reactive substances in rainbow trout fillets by hyperspectral imaging technique

Total phenolic content prediction inFlos Loniceraeusing hyperspectral imaging combined with wavelengths selection methods

Contact Info

Product

Resources

About