Tea plant cultivar identification is normally achieved manually or by spectroscopic, chromatographic, and other methods that are time-consuming and often inaccurate. In this paper, a method for the identification of three tea cultivars with similar leaf morphology is proposed using transfer learning by five pre-trained models: EfficientNet-B0, MobileNetV2, MobileNetV3, MobileViT-S, and ShuffleNetV2. The results showed that the best test accuracy percentages for EfficientNet-B0, MobileNetV2, MobileNetV3, MobileViT-S, and ShuffleNetV2 were 98.33, 99.67, 99.33, 98.67, and 99.00%, respectively. The most lightweight model was ShuffleNetV2, and the fastest combination was ShuffleNetV2 with 112 × 112 image resolution. Considering accuracy, the number of parameters, and floating point operations (FLOPs), MobileNetV2 was not only the most accurate model, but also both lightweight and fast. The present research could benefit both farmers and consumers via identifying tea cultivars without destructive techniques, a factor that would reduce the adulteration of commodity tea.

Across the world, most of the main tea plant cultivars are green-leaf cultivars, but colored-leaf cultivars have become increasingly popular in recent years. In this research, the differences between purple-red and white-yellow tea plant cultivars were compared on biochemical and proteome levels. White-yellow cultivars had significantly high amino acid (AA) content and low polyphenols/amino acid (PP/AA), while purple-red cultivars showed high PP/AA. Comprehensive analysis of all nutritional compounds revealed that most of them showed significant positive correlations, except AA, and that there were significant negative correlations between AA and other compounds. The nutritional compounds of some individual cultivars differed from the average trend of the classification, using color as a criterion. Twenty-one differential proteins were detected in the purple-red and yellow-white cultivars. Among these proteins, there were 16 upregulated proteins and 5 downregulated proteins in purple-red cultivars. Most of these proteins act in the photosynthetic system. This indicated that tea plants with purple-red leaves performed better in photosynthesis than yellow-white tea plant cultivars. The chloroplast development of white-yellow tea plant cultivars was obstructed and may introduce AA accumulation.

The absence of a high-efficiency and stable genetic transformation system has been a critical bottleneck, impeding both functional gene characterization and precision breeding efforts in Camellia sinensis (tea), and browning is the first problem encountered in tissue culture of tea. In this paper, to identify optimal spectral conditions for minimizing browning in tissue culture, we subjected three tea plant cultivars to distinct light quality treatments and conducted comprehensive metabolomic profiling of their phytochemical contents. This study demonstrates that wavelength-specific light treatments can induce reversible modifications in the physicochemical characteristics of tea leaves, effectively reducing the accumulation of flavonoid compounds, including polyphenols, in plant tissues. Notably, tissues subjected to optimized wavelength conditions exhibit superior performance as explant sources for in vitro culture systems, demonstrating significantly lower browning rates. Comparative analysis of 460 nm, 660 nm, and 730 nm irradiation treatments revealed consistent suppression of polyphenol biosynthesis across all examined cultivars. However, the wavelength eliciting maximal reduction showed significant cultivar-dependent variation, indicating genotype-specific photoresponsive regulation of secondary metabolism.