No abstract

Quantifying the phenolic compounds in plants is essential for maintaining the beneficial effects of plants on human health. Existing measurement methods are destructive and/or time consuming. To overcome these issues, research was conducted to develop a non-destructive and rapid measurement of phenolic compounds using hyperspectral imaging (HSI) and machine learning. In this study, the Arabidopsis was used since it is a model plant. They were grown in controlled and various stress conditions (LED lights and drought). Images were captured using HSI in the range of 400–1,000 nm (VIS/NIR) and 900–2,500 nm (SWIR). Initially, the plant region was segmented, and the spectra were extracted from the segmented region. These spectra were synchronized with plants’ total phenolic content reference value, which was obtained from high-performance liquid chromatography (HPLC). The partial least square regression (PLSR) model was applied for total phenolic compound prediction. The best prediction values were achieved with SWIR spectra in comparison with VIS/NIR. Hence, SWIR spectra were further used. Spectral dimensionality reduction was performed based on discrete cosine transform (DCT) coefficients and the prediction was performed. The results were better than that of obtained with original spectra. The proposed model performance yielded R2-values of 0.97 and 0.96 for calibration and validation, respectively. The lowest standard errors of predictions (SEP) were 0.05 and 0.07 mg/g. The proposed model out-performed different state-of-the-art methods. These demonstrate the efficiency of the model in quantifying the total phenolic compounds that are present in plants and opens a way to develop a rapid measurement system.

The increasing interest in plant phenolic compounds in the past few years has become necessary because of their several important physicochemical properties. Thus, their identification through non-destructive methods has become crucial. This study carried out comparative non-destructive measurements of Arabidopsis thaliana leaf powder sample phenolic compounds using Fourier-transform infrared and near-infrared spectroscopic techniques under six distinct stress conditions. The prediction analysis of 600 leaf powder samples under different stress conditions (LED lights and drought) was performed using PLSR, PCR, and NAS-based HLA/GO regression analysis methods. The results obtained through FT-NIR spectroscopy yielded the highest correlation coefficient (Rp2) value of 0.999, with a minimum error (RMSEP) value of 0.003 mg/g, based on the PLSR model using the MSC preprocessing method, which was slightly better than the correlation coefficient (Rp2) value of 0.980 with an error (RMSEP) value of 0.055 mg/g for FT-IR spectroscopy. Additionally, beta coefficient plots present spectral differences and the identification of important spectral signatures sensitive to the phenolic compounds in the measured powdered samples. Thus, the obtained results demonstrated that FT-NIR spectroscopy combined with partial least squares regression (PLSR) and suitable preprocessing method has a solid potential for non-destructively predicting phenolic compounds in Arabidopsis thaliana leaf powder samples.

Ginseng is an important medicinal plant in Korea. The roots of the ginseng plant have medicinal properties; thus, it is very important to maintain the quality of ginseng roots. Root rot disease is a major disease that affects the quality of ginseng roots. It is important to predict this disease before it causes severe damage to the plants. Hence, there is a need for a non-destructive method to identify root rot disease in ginseng plants. In this paper, a method to identify the root rot disease by analyzing the RGB plant images using image processing and deep learning is proposed. Initially, plant segmentation is performed, and then the noise regions are removed in the plant images. These images are given as input to the proposed linear deep learning model to identify root rot disease in ginseng plants. Transfer learning models are also applied to these images. The performance of the proposed method is promising in identifying root rot disease.