The efficiency of hyper- and multispectral imaging (HSI and MSI) has gained considerable attention in research on plant phenotyping. This is due to their ease of use while being considered a nondestructive technology. Unlike current point-scanned spectroscopy, both HSI and MSI extract spatial and spectral information while covering a wide range of a plant body. Moreover, it is necessary to equip the extracted information with multivariate calibration techniques, followed by model evaluation. To date, the application of HSI and MSI for monitoring plant growth under a controlled environment is emerging and showing a good trend. Our systematic literature review discusses spectroscopy imaging techniques and their chemometric approaches as a sustainable sensor technology to detect plant traits. In conclusion, we also explore the possibility of carrying out HSI and MSI during plant trait analysis.

Pork adulteration detection in beef is important due to health, economic, and religious concerns. This study explored the use of a Shortwave Near Infrared–Hyperspectral Imaging (SWNIR–HSI) system which captured spectral data across 894–2504 nm to detect adulteration of pork in beef. In this study, minced pork in various concentrations ranging from 0–50% (w/w) were added to pure minced beef. Spectra obtained from the SWNIR–HSI were used to develop a partial least square regression (PLSR) model. The study compared the PLSR results between full wavelengths (variables) and selected wavelengths obtained via the variable importance in projection (VIP) method. The best results from the full-wavelength PLSR model yielded a prediction accuracy (R²P) of 0.940 and a standard error of prediction (SEP) of 4.633%, while using VIP-selected wavelengths improved performance, with R²P of 0.955 and SEP of 3.811%. The study demonstrates the potency of SWNIR–HIS, particularly with selected wavelengths, as an effective and nondestructive tool for accurately predicting pork adulteration levels in beef.

The interest in traditional meat being replaced by plant-based food has increased throughout the years. Some agricultural products, such as root crops, could be incorporated into alternative meat products due to the health benefits. However, relevant studies have discovered that some roots are considered allergen materials, necessitating further identification to maintain consumer safety. Aside from high accuracy, the limitations offered by traditional identification methods are a reason to employ nondestructive methods. This study aimed to develop a hyperspectral imaging system measuring the 400 nm to 1000 nm spectral range for the nondestructive identification of roots in soybean-based patty. Four thin-sliced traditional medicinal roots (tianma (Gastrodia elata), balloon flower root (Platycodon grandiflorum), deodeok (Codonopsis lanceolata), and ginseng (Panax ginseng)) were incorporated in a soybean-based patty with a concentration of 5% w/w. Moreover, support vector machine (SVM) learning and one-dimensional convolutional neural networks (1D-CNN) were realized for the discrimination model in tandem with spectral data extracted from the hyperspectral image. Our study demonstrated that SVM learning effectively discriminates between original patty and patty with root addition, with an F1-score, precision, and recall beyond 96.77%. This optimum model was achieved by using the standard normal variate (SNV) spectra.

Increasing concern for food safety and quality leads to a growing interest in using edible insects as a new food source. Our research proposed using near-infrared (NIR) spectroscopy to detect chickpea and soybean flour in Protaetia brevitarsis seulensis (PBS) powder. The adulteration was made by adding the chickpea and soybean flour with 5% increments to the PBS powder and the total weight was five grams. Our proposal utilized two regression machine learning techniques: partial least squares regression (PLSR) and Gaussian process regression (GPR). We found that GPR using the original spectra was the most effective model for predicting chickpea content indicated by an R 2 p of 0.996, root mean squared error of prediction (RMSEP) of 0.970%, and ratio of prediction to deviation (RPD) of 15.970, while the optimum model for predicting soybean content had an R 2 p of 991, RMSEP of 1.435%, and RPD of 10.778. This approach showed promise for detecting plant-based adulterants in PBS powder.