TY - JOUR
T1 - Global calibration for non-targeted fraud detection in quinoa flour using portable hyperspectral imaging and chemometrics
AU - Wu, Qianyi
AU - Mousa, Magdi A.A.
AU - Al-Qurashi, Adel D.
AU - Ibrahim, Omer H.M.
AU - Abo-Elyousr, Kamal A.M.
AU - Rausch, Kent
AU - Abdel Aal, Ahmed M.K.
AU - Kamruzzaman, Mohammed
N1 - All the imaging data were acquired using the Specim IQ hyperspectral imaging system in reflectance mode. The system consists of a battery-based portable 12-bit Specim V10E IQ line-scan hyperspectral imaging camera with a CMOS sensor, a lens, and an imaging spectrograph (Specim, Spectral Imaging Ltd, Oulu, Finland); an illumination set with two lamps as the light source (Oulu, Finland); a tripod for supporting and fixing the camera at a constant height; and an output analyzing computer with Specim IQ Studio software installed for extracting the image and spectral information. Each scan of the system produces 204 spectral bands in the wavelength range between 400 and 1000 nm as well as a 2D square image with a resolution of 512 × 512 pixels. As a result, a 3D hypercube was created with dimensions of 512 × 512 × 204 with over 53 million data points for each scan. The system was operated in a completely dark environment to ensure that the only light sources were from the pre-set illumination lamps so that the effect of unintended stray light could be eliminated. The camera was fixed on the tripod facing down, while samples were placed on a blackboard below the camera for scanning. The blackboard was put on a flat surface with a fixed height to maintain a uniform distance from the camera to the samples throughout the data gathering process. Before the data acquisition process, a random sample on the blackboard with a white reference panel (∼99% reflectance) was scanned for system calibration. This calibration data were saved and used for the reflectance transformation for every scan taken for the samples. After proper system calibration, the prepared samples were put into a small black container, approximately 4 cm in diameter and 2 cm in depth. One sample was put exactly below the camera for image acquisition. The dark reference data (∼0% reflectance) were captured inside the system automatically by blocking the incoming light completely. The transformation can be symbolized in the following equation:The authors extend their appreciation to the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the project number “IFPRC: 188–155-2020″ and King Abdulaziz University, DSR, Jeddah, Saudi Arabia.
The authors extend their appreciation to the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the project number “ IFPRC: 188--155-2020 ″ and King Abdulaziz University , DSR , Jeddah, Saudi Arabia.
PY - 2023/1
Y1 - 2023/1
N2 - Quinoa is one of the highest nutritious grains, and global consumption of quinoa flour has increased as people pay more attention to health. Due to its high value, quinoa flour is susceptible to adulteration. Cross-contamination between quinoa flour and other flour can be easily neglected due to their highly similar appearance. Therefore, detecting adulteration in quinoa flour is important to consumers, industries, and regulatory agencies. In this study, portable hyperspectral imaging in the visible near-infrared (VNIR) spectral range (400–1000 nm) was applied as a rapid tool to detect adulteration in quinoa flour. Quinoa flour was adulterated with wheat, rice, soybean, and corn in the range of 0–98% with 2% increments. Partial least squares regression (PLSR) models were developed, and the best model for detecting the % authentic flour (quinoa) was obtained by the raw spectral data with R2p of 0.99, RMSEP of 3.08%, RPD of 8.77, and RER of 25.32. The model was improved, by selecting only 13 wavelengths using bootstrapping soft shrinkage (BOSS), to R2p of 0.99, RMSEP of 2.93%, RPD of 9.18, and RER of 26.60. A visualization map was also generated to predict the level of quinoa in the adulterated samples. The results of this study demonstrate the ability of VNIR hyperspectral imaging for adulteration detection in quinoa flour as an alternative to the complicated traditional method.
AB - Quinoa is one of the highest nutritious grains, and global consumption of quinoa flour has increased as people pay more attention to health. Due to its high value, quinoa flour is susceptible to adulteration. Cross-contamination between quinoa flour and other flour can be easily neglected due to their highly similar appearance. Therefore, detecting adulteration in quinoa flour is important to consumers, industries, and regulatory agencies. In this study, portable hyperspectral imaging in the visible near-infrared (VNIR) spectral range (400–1000 nm) was applied as a rapid tool to detect adulteration in quinoa flour. Quinoa flour was adulterated with wheat, rice, soybean, and corn in the range of 0–98% with 2% increments. Partial least squares regression (PLSR) models were developed, and the best model for detecting the % authentic flour (quinoa) was obtained by the raw spectral data with R2p of 0.99, RMSEP of 3.08%, RPD of 8.77, and RER of 25.32. The model was improved, by selecting only 13 wavelengths using bootstrapping soft shrinkage (BOSS), to R2p of 0.99, RMSEP of 2.93%, RPD of 9.18, and RER of 26.60. A visualization map was also generated to predict the level of quinoa in the adulterated samples. The results of this study demonstrate the ability of VNIR hyperspectral imaging for adulteration detection in quinoa flour as an alternative to the complicated traditional method.
KW - Adulteration
KW - BOSS
KW - PLSR
KW - Quinoa flour
KW - VNIR hyperspectral imaging
KW - Visualization
UR - https://www.scopus.com/pages/publications/85151045997
UR - https://www.scopus.com/inward/citedby.url?scp=85151045997&partnerID=8YFLogxK
U2 - 10.1016/j.crfs.2023.100483
DO - 10.1016/j.crfs.2023.100483
M3 - Article
C2 - 37033735
AN - SCOPUS:85151045997
SN - 2665-9271
VL - 6
JO - Current Research in Food Science
JF - Current Research in Food Science
M1 - 100483
ER -