Development of Noninvasive Classification Methods for Different Roasting Degrees of Coffee Beans Using Hyperspectral Imaging

This study aimed to develop an approach for quickly and noninvasively differentiating the roasting degrees of coffee beans using hyperspectral imaging (HSI). The qualitative properties of seven roasting degrees of coffee beans (unroasted, light, moderately light, light medium, medium, moderately dark, and dark) were assayed, including moisture, crude fat, trigonelline, chlorogenic acid, and caffeine contents. These properties were influenced greatly by the respective roasting degree. Their hyperspectral images (874–1734 nm) were collected using a hyperspectral reflectance imaging system. The spectra of the regions of interest were manually extracted from the HSI images. Then, principal components analysis was employed to compress the spectral data and select the optimal wavelengths based on loading weight analysis. Meanwhile, the random frog (RF) methodology and the successive projections algorithm were also adopted to pick effective wavelengths from the spectral data. Finally, least squares support vector machine (LS-SVM) was utilized to establish discriminative models using spectral reflectance and corresponding labeled classes for each degree of roast sample. The results showed that the LS-SVM model, established by the RF selecting method, with eight wavelengths performed very well, achieving an overall classification accuracy of 90.30%. In conclusion, HSI was illustrated as a potential technique for noninvasively classifying the roasting degrees of coffee beans and might have an important application for the development of nondestructive, real-time, and portable sensors to monitor the roasting process of coffee beans.

[1]  Leandro S. Oliveira,et al.  A preliminary evaluation of the effect of processing temperature on coffee roasting degree assessment , 2009 .

[2]  Nicola Caporaso,et al.  Non-destructive analysis of sucrose, caffeine and trigonelline on single green coffee beans by hyperspectral imaging , 2018, Food research international.

[3]  Siwalak Pathaveerat,et al.  Classification of longan fruit bruising using visible spectroscopy , 2011 .

[4]  Draženka Komes,et al.  Comparative study of polyphenols and caffeine in different coffee varieties affected by the degree of roasting. , 2011, Food chemistry.

[5]  Nicola Caporaso,et al.  Rapid prediction of single green coffee bean moisture and lipid content by hyperspectral imaging , 2018, Journal of food engineering.

[6]  Chu Zhang,et al.  Variety Identification of Single Rice Seed Using Hyperspectral Imaging Combined with Convolutional Neural Network , 2018 .

[7]  Ajmal Mian,et al.  Using hyperspectral imaging to characterize consistency of coffee brands and their respective roasting classes , 2016 .

[8]  A A Gowen,et al.  Characterisation of hydrogen bond perturbations in aqueous systems using aquaphotomics and multivariate curve resolution-alternating least squares. , 2013, Analytica chimica acta.

[9]  B. Bakshi.  Multiscale PCA with application to multivariate statistical process monitoring , 1998 .

[10]  Yong He,et al.  Mapping of Chlorophyll and SPAD Distribution in Pepper Leaves During Leaf Senescence Using Visible and Near-Infrared Hyperspectral Imaging , 2016 .

[11]  Felix Escher,et al.  Coffee roasting and aroma formation: application of different time-temperature conditions. , 2008, Journal of agricultural and food chemistry.

[12]  Qing-Song Xu,et al.  Random frog: an efficient reversible jump Markov Chain Monte Carlo-like approach for variable selection with applications to gene selection and disease classification. , 2012, Analytica chimica acta.

[13]  Chiara Cordero,et al.  Non-separative headspace solid phase microextraction-mass spectrometry profile as a marker to monitor coffee roasting degree. , 2013, Journal of agricultural and food chemistry.

[14]  Isobel Davidson,et al.  The effect of green-coffee-bean extract rich in chlorogenic acid on antioxidant status of healthy human volunteers , 2010, Proceedings of the Nutrition Society.

[15]  Da-Wen Sun,et al.  Recent Advances in Wavelength Selection Techniques for Hyperspectral Image Processing in the Food Industry , 2014, Food and Bioprocess Technology.

[16]  J. Roger,et al.  Application of LS-SVM to non-linear phenomena in NIR spectroscopy: development of a robust and portable sensor for acidity prediction in grapes , 2004 .

[17]  Antonio J. Plaza,et al.  A fast iterative algorithm for implementation of pixel purity index , 2006, IEEE Geoscience and Remote Sensing Letters.

[18]  M. Aparecida,et al.  Optimization of the roasting of robusta coffee (C. canephora conillon) using acceptability tests and RSM , 2001 .

[19]  Fei Liu,et al.  Application of Visible and Near Infrared Hyperspectral Imaging to Differentiate Between Fresh and Frozen–Thawed Fish Fillets , 2013, Food and Bioprocess Technology.

[20]  Pengcheng Nie,et al.  Application of Time Series Hyperspectral Imaging (TS-HSI) for Determining Water Distribution Within Beef and Spectral Kinetic Analysis During Dehydration , 2013, Food and Bioprocess Technology.

[21]  Santina Romani,et al.  Near infrared spectroscopy: an analytical tool to predict coffee roasting degree. , 2008, Analytica chimica acta.

[22]  Eulália Mendes,et al.  Roast effects on coffee amino acid enantiomers , 2005 .

[23]  Roman M. Balabin,et al.  Support vector machine regression (SVR/LS-SVM)--an alternative to neural networks (ANN) for analytical chemistry? Comparison of nonlinear methods on near infrared (NIR) spectroscopy data. , 2011, The Analyst.

[24]  Bart Nicolai,et al.  Prediction of optimal cooking time for boiled potatoes by hyperspectral imaging , 2011 .

[25]  Leandro S. Oliveira,et al.  A preliminary study on the feasibility of using the composition of coffee roasting exhaust gas for the determination of the degree of roast , 2001 .

[26]  William J. Welch,et al.  Computer-aided design of experiments , 1981 .

[27]  Chu Zhang,et al.  Application of Near-Infrared Hyperspectral Imaging with Variable Selection Methods to Determine and Visualize Caffeine Content of Coffee Beans , 2016, Food and Bioprocess Technology.

[28]  João Rodrigo Santos,et al.  In-line monitoring of the coffee roasting process with near infrared spectroscopy: Measurement of sucrose and colour. , 2016, Food chemistry.

[29]  Gamal ElMasry,et al.  Non-destructive prediction and visualization of chemical composition in lamb meat using NIR hyperspectral imaging and multivariate regression , 2012 .

[30]  Hu Shuang-fan.  Correlation analysis between chemical components and sensory quality of coffee , 2013 .

[31]  Takayuki Shibamoto,et al.  Chlorogenic acid and caffeine contents in various commercial brewed coffees , 2008 .

[32]  João Rodrigo Santos,et al.  A Non-invasive Real-Time Methodology for the Quantification of Antioxidant Properties in Coffee During the Roasting Process Based on Near-Infrared Spectroscopy , 2017, Food and Bioprocess Technology.

[33]  Gamal ElMasry,et al.  Fast detection and visualization of minced lamb meat adulteration using NIR hyperspectral imaging and multivariate image analysis. , 2013, Talanta.

[34]  Noel D.G. White,et al.  Extraction of Spectral Information from Hyperspectral Data and Application of Hyperspectral Imaging for Food and Agricultural Products , 2016, Food and Bioprocess Technology.

[35]  M. C. U. Araújo,et al.  The successive projections algorithm for variable selection in spectroscopic multicomponent analysis , 2001 .

[36]  Stéphane Dussert,et al.  Caffeine, trigonelline, chlorogenic acids and sucrose diversity in wild Coffea arabica L. and C. Canephora P. accessions , 2001 .

[37]  Byoung-Kwan Cho,et al.  Qualitative properties of roasting defect beans and development of its classification methods by hyperspectral imaging technology. , 2017, Food chemistry.