Tea plant (Camellia sinensis) is one of the most significant economic crops in modern agriculture. Anthracnose is one of the most destructive fungal diseases in tea plants. It is often required for its early diagnosis and dynamic monitoring for the tea yield and quality. However, it is still lacking on the early in situ diagnosis of anthracnose at the leaf scale at present. This study aims to realize the early diagnosis of tea anthracnose using hyperspectral imaging. Two tea cultivars, “Zhongcha 108” and “Longjing 43”, were utilized as the experimental materials. Four hormones, six defense enzymes, and three photosynthetic pigments were then determined in the anthracnose-resistant and susceptible varieties. Hyperspectral imaging was employed to capture the leaf images at five stages of the anthracnose infection (0, 12, 24, 36 and 72 h). The spectral response features of the tea leaves were obtained from the two varieties at different infection stages. Principal component analysis (PCA) was employed on the hyperspectral data of the tea leaves using unsupervised clustering. Furthermore, the dynamic monitoring models were established for the hormones, defense enzymes, and photosynthetic pigments of the tea leaves. Additionally, the vertex component analysis (VCA) algorithm was used to unmix the hyperspectral image of the tea leaves. In hormones, there were the comparable concentrations of the salicylic acid (SA) and abscisic acid (ABA) between the two cultivars. The similar increasing trends were found at 24 h post-inoculation (hpi) and peaking after 72 h. In contrast, the pronounced differences were observed in the content of jasmonic acid (JA) and indole-3-acetic Acid (IAA). The JA content in “Zhongcha 108” increased at a higher rate, compared with the “Longjing 43”. Furthermore, the IAA content in “Zhongcha 108” was consistently remained more than double that in “Longjing 43” in the infection period (except at 0 hpi). These divergent hormonal responses were likely associated with the differential resistance to anthracnose between the two cultivars. The activities of five defense enzymes (peroxidase (POD), superoxide dismutase (SOD), malondialdehyde (MDA), phenylalamine ammonia lyase (PAL), and polyphenol oxidase (PPO)) in both cultivars increased with the duration of infection, thereby reaching their maximum levels after 72 hpi. Notably, the PPO activity in the “Zhongcha 108” was higher than that in the “Longjing 43”. Additionally, the catalase (CAT) activity in the “Zhongcha 108” displayed an upward trend, whereas it declined in the “Longjing 43”. Therefore, the PPO and CAT also played significant roles in the tea plant's resistance to the anthracnose. The contents of the photosynthetic pigments in two cultivars, including chlorophyll a, chlorophyll b, and carotenoids, decreased progressively with the extension of the infection time, thus reaching their minimum after 72 h. In hyperspectral imaging, The spectral features (peak and valley positions) were observed between the two tea varieties at different infection stages. There was the dynamic variation in the component contents of the tea leaves. Clustering results showed that the samples with the degree of infection at each stage were fully identified in the principal component space (cumulative contribution rate > 96%). Partial least squares regression (PLSR) was used to establish a quantitative model between the average spectrum of the tea leaves and physiological and biochemical indicators, with the maximum correlation coefficient of 0.8924. The number of model variables was reduced from 288 to 10 after feature wavelength selection (competitive adaptive reweighted sampling, CARS). The performance of most quantitative models was improved after selection. The spectral unmixing was greatly contributed to the in-situ visualization of the spatiotemporal dynamics of the disease lesions at the pixel scale, particularly for the early diagnosis of anthracnose 12 h after inoculation. There was 12~24 h earlier than the polymerase chain reaction (PCR). This finding can provide the technical support for the disease prevention and control in tea gardens. The perspective can also offer for the interaction between plants and fungal diseases.
- Article type
- Year
- Co-author
Tea polyphenols, as a key indicator for evaluating tea quality, possess significant health benefits. Traditional detection methods are limited by poor timeliness, high cost, and destructive sampling, making them difficult to meet the demands of tea cultivar breeding and real-time monitoring of tea quality. Meanwhile, rapid identification of tea cultivars and leaf positions is critical for guiding tea production. Therefore, this study aims to develop a non-destructive detection device for quality components of fresh tea leaves based on the combined technology of visible/short-wave near-infrared and long-wave near-infrared spectroscopy, to realize rapid non-destructive detection of tea polyphenol content and rapid identification of tea cultivars and leaf positions.
A rapid non-destructive detection device for quality components of fresh tea leaves was developed by combining visible/short-wave near-infrared spectroscopy (400~1050 nm) and long-wave near-infrared spectroscopy (1051~1650 nm). The Savitzky-Golay (SG) convolution smoothing method was used for preprocessing the spectral data. The Folin-Ciocalteu method was employed to determine the tea polyphenol content, and abnormal samples were eliminated using the interquartile range (IQR) method. Data-level and feature-level fusion methods were adopted, with the competitive adaptive reweighted sampling (CARS) algorithm used to extract characteristic wavelengths. Prior to modeling, the Kennard-Stone algorithm was applied to partition the dataset into a training set and a prediction set at a ratio of 4∶1. Models such as principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), least squares support vector machine (LS-SVM), extreme learning machine (ELM), and 1D convolutional neural network (1D-CNN) were constructed for the identification of 3 cultivars (Huangdan, Tieguanyin, and Benshan) and 4 leaf positions. For predicting tea polyphenol content, models including partial least squares regression (PLSR), least squares support vector regression (LS-SVR), ELM, and 1D-CNN were established for predicting the tea polyphenol content in fresh tea leaves.
The results showed that there were significant differences in tea polyphenol contents among different cultivars and leaf positions (P < 0.05). Specifically, the tea polyphenol content of Huangdan was 17.54%±1.82%, which was 1.16 times and 1.04 times that of Tieguanyin (15.04%±1.22%) and Benshan (16.81%±1.24%), respectively. For each cultivar, the tea polyphenol content generally showed a decreasing trend from the 1st to 4th leaf positions, with the highest content in the 1st leaf position. Principal component analysis (PCA) revealed that for cultivar identification, the scatter distribution of the principal components of Huangdan, Tieguanyin, and Benshan, as well as their projections in the directions of PC1 and PC2, showed a clear trend of clustering into three groups, indicating a good classification effect, although there was still some overlap among individual samples. For leaf position identification, the scatter distributions of the principal components of the 1st, 2nd, 3rd, and 4th leaf positions overlapped with each other, with no obvious clustering among leaf positions. Compared with single-source data, models based on data fusion effectively improved prediction performance. Among them, the PLS-DA model established by combining SG preprocessing with feature-level fusion achieved prediction accuracies of 100% and 87.93% for the identification of 3 tea cultivars and 4 leaf positions, respectively. Furthermore, the 1D-CNN model based on data-level fusion exhibited superior performance in predicting tea polyphenol content, with a coefficient of determination (RP2), root mean square error of prediction (RMSEP), and residual predictive deviation (RPD) of 0.8020, 0.6368%, and 2.2684, respectively, which outperformed models using only visible/short-wave near-infrared spectroscopy or long-wave near-infrared spectroscopy.
The developed detection device combining visible/short-wave near-infrared and long-wave near-infrared spectroscopy, mainly composed of spectrometers, Y-type optical fibers, plant probes, polymer lithium batteries, DC uninterruptible power supplies, voltage conversion modules, and aluminum alloy casings, could synchronously collect multi-source spectral data of visible/short-wave near-infrared and long-wave near-infrared from fresh tea leaves. Combined with data fusion methods and machine learning algorithms, it enabled rapid detection of tea polyphenol content and efficient identification of cultivars and leaf positions in fresh tea leaves, providing new insights for the application of multi-source data fusion technology in elite tea cultivar breeding and non-destructive detection of fresh tea leaf quality.
京公网安备11010802044758号