Near-infrared spectroscopy and multivariate calibration as an alternative to the Agtron to predict roasting degrees in coffee beans and ground coffees

Pires, Fabiana de Carvalho; Pereira, Rosemary Gualberto Fonseca Alvarenga; Baqueta, Michel Rocha; Valderrama, Patrícia; Rocha, Roney Alves da

doi:10.1016/j.foodchem.2021.130471

Cited by 30 publications

(17 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After drying, 8 kg of dry parchment coffee (DPC) was obtained; this mass was then hulled to remove the endocarp (parchment), from which 720 g were peaberries. Then, the green coffee was roasted in a PROBAT sample roaster, four 150 g samples of peaberries and four 150 g samples of standard beans were used for the cupping analysis while the rest of the coffee was used for the different tests, the initial drum temperature was 200°C, and the samples were roasted during 12 minutes at a constantly increasing temperature 25°C – 225°C until reaching a medium‐high roasting profile with a color level of 55 agtron (Pires et al, 2021), the medium‐high roasting degree was chosen to detect easier and see the roasting tonality differences due to the contrast generated between the colours (Darmajana et al, 2022; Odzakovic et al, 2019).…”

Section: Methodsmentioning

confidence: 99%

Peaberry shape and size influence on different coffee postharvest processes

Duque‐Dussán,

Figueroa‐Varela,

Sanz‐Uribe

2023

J Food Process Engineering

View full text Add to dashboard Cite

In many coffee‐producing countries, the ellipsoidal‐shaped seeds called peaberries are often labeled as a defect because of their shape and reduced size, going against the market demand for large‐sized standard coffee beans. Nevertheless, the peaberry natural occurrence on the coffee plantations is significant, accounting for 5%–7% of the total harvested coffee for Coffea arabica L. and Coffea canephora, the most planted species worldwide. Nevertheless, recent growth in the peaberry market has happened due to exceptional cupping scores for this specific bean type; however, the relationship between these scores and the shape of the bean was not yet recorded in the literature. Therefore, this research aimed to evaluate and compare the impact of the shape and size of the peaberry against the standard beans in different postharvest processes: Drying, roasting (colorimetry and inner roasting profiles), grinding (compressive and shear force tests) and overall quality by cupping analysis. Coffea arabica L. var. Cenicafé 1 was used throughout all the experiments, where advanced methods were used to increase the accuracy of the results and deeply characterize the process behavior. The results of this research allow to understand the peaberry postharvest behavior better and add significant value to this often‐underrated bean condition. The peaberries demonstrated shape influence in the different evaluated parameters, allowing them to dry faster, roast evenly, avoid burnt spots, and collapse at homogeneous forces while attaining the same high cup scores as a standard coffee bean.Practical applicationsUnderstanding the peaberries' shape and size influence different postharvest processes is crucial to comprehend their value. By gaining insights, into how these unique beans interact with stages of processing, producers can control and predict their behavior leading to more consistent and optimized results. The findings of this research provide an opportunity to fully incorporate peaberries into the coffee product without considering them as defective beans or expanding market trends. Generally, peaberries were often seen as undesirable because they did not match the bean size and shape. However, their exceptional performance in drying, roasting, grinding and cupping dispels these misconceptions, highlighting their value and contribution to coffee quality. As a result, coffee producers can embrace peaberries as valuable beans without discarding or downgrading them. This newfound appreciation for peaberries does not only reduce waste, but it also diversifies coffee offerings to meet consumer preferences and enriches the industry.

show abstract

Section: Methodsmentioning

confidence: 99%

Peaberry shape and size influence on different coffee postharvest processes

Duque‐Dussán,

Figueroa‐Varela,

Sanz‐Uribe

2023

J Food Process Engineering

View full text Add to dashboard Cite

show abstract

“…This study tested modeling by leaving one cross-validation model with ratio data calibration to validation around 80%:10%. The model's interpretation was determined using the coefficient correlation of calibration and validation, root means square error of calibration (RMSEC), and root means square error of prediction (RMSEP) as specified by Pires et al [16]. Additionally, the error difference between RMSEC and RMSEP and the ratio of prediction to deviation (RPD) were analyzed using the Caporaso et al [17] equation.…”

Section: Calibration and Validation Modelmentioning

confidence: 99%

Performance Evaluation of Pre-Processing and Pre-Treatment Algorithm for Near-Infrared Spectroscopy Signals: Case Study pH of Intact Mango “Arumanis”

Agustina¹,

Devianti²,

Bulan³

et al. 2022

IJDNE

View full text Add to dashboard Cite

pH is one of the important physical parameters to characterize mango damage because it can indicate changes in the structure and chemical content of the fruit. Thus, the present work evaluated the possibility of NIRs as a rapid and non-destructive tool for measuring the pH properties of intact mango from the cultivar "Arumanis" (Mangifera indica L.) using several algorithms for pre-processing, pre-treatment, and prediction. Three different algorithm predictions, namely principal component regression (PCR), partial least squares regression (PLSR), and support vector machine regression (SVMR), were used and compared to predict the pH of mangos. A total of 16 pre-processing and pre-treatment algorithms are used to support algorithm prediction, and the results are also compared with the raw data spectra. The NIR spectral data used range from 1000 to 2500 nm. Algorithm performance will be evaluated using RMSE, error differences and concluded using RPD. The results show that the prediction of the PLSR algorithm can be performed with an RPD of 8.17, which is more significant than the PCR and SVMR algorithms, which are 1.04, and 1.90, respectively. To support this, pre-processing and pretreatment of the second derivative Savitzky–Golay is the best algorithm that can be used to predict the pH of the whole mango cultivar "Arumanis".

show abstract

“…Unlike direct measurement techniques used in chemical analysis, spectral analysis serves as an indirect method. It necessitates the integration with chemometric methods to construct a model that captures the correlation between the sample and its spectral data [13]. At present, the predominant chemometric methods include Support Vector Machines (SVM), Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (Xgboost), Random Forest (RF), among others.…”

Section: Introductionmentioning

confidence: 99%

Discrimination of small sample tea varieties based on convolutional neural network and deep convolutional generative adversarial network enhanced near-infrared diffuse reflectance spectral dataset

Guo,

Huang,

Sheng

et al. 2024

Preprint

View full text Add to dashboard Cite

Near-infrared diffuse reflectance spectroscopy is widely recognized as a rapid, non-destructive, and environmentally friendly detection technology. However, in order to ensure the accuracy and stability of the detection model, a large number of sample data is necessary. This paper proposed the rapid and non-destructive detection of small sample tea variety recognition based on the near-infrared diffuse reflectance spectrum data extended by convolutional neural network (CNN) and deep convolutional generative adversarial network (DCGAN). The near-infrared diffuse reflectance spectra of 240 tea samples were obtained by Lambda 950 spectrometer using eight of the most popular tea varieties on the Chinese market. Firstly, the spectral data was enhanced using translation, linear superposition, noise addition, and DCGAN methods, and the quality of the generated spectra was evaluated using the support vector machine (SVM) and gradient boosting decision tree (GBDT) methods. Compared with other methods, the DCGAN has the highest accuracy of 91.75%. Secondly, the optimal number of iterations of DCGAN was confirmed to be 6000 by SVM and GBDT methods. To further augment the precision of identifying small samples of tea, two additional classification models of the Extreme Gradient Boosting (Xgboost) and CNN were applied to the DCGAN. Finally, the results demonstrated that the CNN achieved the highest identification accuracy of 98.68% compared with SVM (90.46%), GBDT (90.42%), and Xgboost (88.83%) with an additional 100 samples and 6000 iterations. Therefore, the combination of deep convolutional generative adversarial network enhanced near-infrared diffuse reflectance spectral dataset and the CNN successfully realizes the identification of small sample tea varieties. The experimental results strongly indicate that this method holds significant potential for practical implementation in the field of small sample tea varieties identification.

show abstract

Near-infrared spectroscopy and multivariate calibration as an alternative to the Agtron to predict roasting degrees in coffee beans and ground coffees

Cited by 30 publications

References 33 publications

Peaberry shape and size influence on different coffee postharvest processes

Peaberry shape and size influence on different coffee postharvest processes

Performance Evaluation of Pre-Processing and Pre-Treatment Algorithm for Near-Infrared Spectroscopy Signals: Case Study pH of Intact Mango “Arumanis”

Discrimination of small sample tea varieties based on convolutional neural network and deep convolutional generative adversarial network enhanced near-infrared diffuse reflectance spectral dataset

Contact Info

Product

Resources

About