The use of multispectral imaging for the discrimination of Arabica and Robusta coffee beans

Mihailova, Alīna; Liebisch, Beatrix; Islam, Marivil; Carstensen, Jens Michael; Cannavan, Andrew; Kelly, Simon

doi:10.1016/j.fochx.2022.100325

Cited by 15 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We then validated the color recognition and discrimination performance of the proposed multispectral vision system using the samples. Compared with Leme et al (2019), Mihailova et al (2022), this system requires less extensive training and is less computationally intensive, thereby reducing hardware costs. Additionally, this system conducts tests using unground coffee beans, preserving sample structure.…”

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

“…For the final validation, we conducted an experiment on the classification of adulterated coffee beans, with reference to Leme et al (2019), Mihailova et al (2022), and performed a comparative analysis of the experimental results. We divided coffee beans by type into eight samples of 20 beans each.…”

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

“…Subsequently, we employed these 21 sample combinations to validate the color recognition and discrimination performance of the proposed multispectral vision system. First, we prepared 90-mm-diameter Petri dishes, similar to those mentioned in Mihailova et al (2022). Each Petri dish contained 20 coffee beans, and we selectively replaced 0-20 beans with different roast levels.…”

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

See 2 more Smart Citations

Intelligent classifier for various degrees of coffee roasts using smart multispectral vision system

Lin,

Chen,

2024

Journal of Field Robotics

View full text Add to dashboard Cite

This study proposes an innovative deep learning model for use in a multispectral vision system comprising a complementary metal‐oxide semiconductor image sensor and a spectrometer. To ensure accurate color recognition, the deep learning model includes an embedded adaptive automatic color temperature correction engine. By using this color temperature correction engine, the multispectral vision system can intelligently compensate for lighting and chromatic variations. To evaluate the performance of the system, we created a nine‐dimensional data set using the IT8.7/2 color target. We then used this data set to train the deep learning model. Our deep learning model outperformed other lightweight deep learning models in experiments, making it suitable for deployment on edge devices and embedded systems. We tested the ability of the multispectral vision system to classify adulterated coffee beans into their respective classes. The overall accuracy rate was more than 99.3%, indicating that out proposed multispectral vision system is effective in identifying color differences. Considering its capabilities in agricultural screening, we suggest incorporating our adaptive automatic multispectral vision system into agricultural machines for the realization of Agriculture 4.0.

show abstract

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

Section: Classification Of Adulterated Coffee Beans In Various Degree...mentioning

confidence: 99%

See 1 more Smart Citation

Intelligent classifier for various degrees of coffee roasts using smart multispectral vision system

Lin,

Chen,

2024

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…Coffea arabica (Arabica) and Coffea canephora (Robusta) are the two main coffee species cultivated for commercially producing coffee. Coffea arabica accounts for more than 60% of global coffee production, leading to more byproducts from processed Arabica coffee cherry (Mihailova et al., 2022). Coffee pulp is the main byproduct of coffee fruits after wet processing (Klingel et al., 2020).…”

Section: Introductionmentioning

confidence: 99%

Effects of roasting on the chemical compositions, color, aroma, microstructure, and the kinetics of changes in coffee pulp

Sun

Huang

et al. 2023

Journal of Food Science

View full text Add to dashboard Cite

Roasting is a critical process that affects the quality attributes of coffee beans; however, how roasting conditions affect the physical, chemical, biological, and organoleptic changes of coffee pulp needs more research. In the present study, we investigated the effects of roasting temperatures and times on chemical compositions and quality attributes of coffee pulp. The results showed that the contents of total soluble sugar (TSS) and free amino acid (FAA) followed a temporal pattern of first increasing and then decreasing under the roasting temperatures between 100 and 160 • C. Total phenolic content (TPC) and antioxidant activity of coffee pulp significantly (p < 0.05) increased after roasting, reaching the maximum values of 83.09 mg gallic acid equivalent (GAE) /g and 360.45 µM 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) /g, respectively, when coffee pulp was roasted at 160 • C for 18 min. Drying rates of coffee pulp fitted the Logarithmic kinetic model, while color (L*, a*, and b*) changes and 5-caffeoylquinic acid degradation followed the first-order kinetic model. Electronic nose analysis showed that the main aroma compounds of the coffee pulp are sulfur-containing organics that were reduced with the extended roasting time. Scanning electronic microscopy analysis presented the loosened, shrunk, and cracked microstructure on the surface of the roasted coffee pulp, which might contribute to the increased TSS, FAA, TPC, and antioxidant activity of coffee pulp roasted under specific conditions. In conclusion, our research provides valuable information for preparing high-quality coffee pulp tea.

show abstract

“…Tere are three types of cofee in the market: green, roasted, and processed cofee [20], all are susceptible to being adulterated with the by-products of cofee or other plant materials, such as cofee husks [21,22], waste cofee grounds [23], brown sugar [24], acai [25], cheaper grains (wheat [26], rice [27], rye [28], barley [2,29,30], corn [26,[31][32][33], soybean [32]), and chicory [30,34]. Besides, intentionally mislabeling lower-valued beans as their higher-valued counterparts and blending different species or origins of beans are other possible forms of fraud [35,36]. In response to the increasing public concerns of fraudulent practices about cofee, the consumers and actors of the food industry sector, including the related government agencies and food companies, demand the methods for the characterization of cofee to be adequately reliable, accurate, high-throughput, low-cost, as well as easy-to-operate with minimal human intervention [37].…”

Section: Introductionmentioning

confidence: 99%

Recent Research Advancements of Coffee Quality Detection: Targeted Analyses vs. Nontargeted Fingerprinting and Related Issues

Chen

Gao

2023

Journal of Food Quality

View full text Add to dashboard Cite

Coffee is an important commercial product that arose various quality issues. Different techniques have been applied to detect coffee quality. This review focused on the recent updates in the detection methods of coffee from a targeted versus nontargeted perspective. This review introduced case studies of the current research progresses on targeted and nontargeted detection approaches. Their merits and demerits were evaluated as an analysis of coffee quality. The targeted approach, including liquid chromatography (LC), gas chromatography (GC), thin-layer chromatography (TLC), and capillary electrophoresis (CE), evaluates the quality of coffee by specific markers, whereas the nontargeted approach tests whether the sample is abnormal, without prior knowledge of what caused the abnormality, usually coupled with chemometrics. The nontargeted techniques commonly involve LC, GC, nuclear magnetic resonance (NMR), infrared spectroscopy (IR), ultraviolet-visible spectroscopy (UV-Vis), laser-induced breakdown spectroscopy (LIBS), and mass spectrometry (MS). This work may provide guidance for resolving most aspects of the quality problems in coffee, such as adulterant detection, species identification, and geographical origin discrimination.

show abstract

The use of multispectral imaging for the discrimination of Arabica and Robusta coffee beans

Cited by 15 publications

References 24 publications

Intelligent classifier for various degrees of coffee roasts using smart multispectral vision system

Intelligent classifier for various degrees of coffee roasts using smart multispectral vision system

Effects of roasting on the chemical compositions, color, aroma, microstructure, and the kinetics of changes in coffee pulp

Recent Research Advancements of Coffee Quality Detection: Targeted Analyses vs. Nontargeted Fingerprinting and Related Issues

Contact Info

Product

Resources

About