No abstract

The roots of Astragalus membranaceus var. mongholicus (AMM) and A. membranaceus (AM) are widely used in traditional Chinese medicine. Although AMM has higher yields and accounts for a larger market share, its cultivation is fraught with challenges, including mixed germplasm resources and widespread adulteration of commercial seeds. Current methods for distinguishing Astragalus seeds from similar (SM) seeds are time-consuming, laborious, and destructive. To establish a non-destructive method, AMM, AM, and SM seeds were collected from various production areas. Machine vision and hyperspectral imaging (HSI) were used to collect morphological data and spectral data of each seed batch, which was used to establish discriminant models through various algorithms. Several preprocessing methods based on hyperspectral data were compared, including multiplicative scatter correction (MSC), standard normal variable (SNV), and first derivative (FD). Then selection methods for identifying informative features in the above data were compared, including successive projections algorithm (SPA), uninformative variable elimination (UVE), and competitive adaptive reweighted sampling (CARS). The results showed that support vector machine (SVM) modeling of machine vision data could distinguish Astragalus seeds from SM with >99% accuracy, but could not satisfactorily distinguish AMM seeds from AM. The FD-UVE-SVM model based on hyperspectral data reached 100.0% accuracy in the validation set. Another 90 seeds were tested, and the recognition accuracy was 100.0%, supporting the stability of the model. In summary, HSI data can be applied to discriminate among the seeds of AMM, AM, and SM non-destructively and with high accuracy, which can drive standardization in the Astragalus production industry.

Seed processing is an important means of improving seed quality. However, the traditional seed processing process and parameter adjustment are highly empirically dependent. In this study, machine vision technology was used to develop a seed processing method based on the rapid extraction of seeds’ material characteristics. Combined with the results of clarity analysis and the single seed germination test, the seed processing process and parameters were determined through data analysis. The results showed that several phenotypic features were significantly or highly significantly correlated with clarity, but fewer phenotypic features were correlated with viability. According to the probability density distribution of pure seeds and impurities in the features that were significantly correlated with seed clarity, the sorting parameters of length, width, R, G, and B were determined. When the combination of width (≥0.8 mm) + G (<75) was used for sorting, the recall of pure seeds was higher than 91%, and the precision was increased to 98.6%. Combined with the specific production reality, the preliminary determination of the Platycodon grandiflorum seed processing process was air separation—screen (round hole sieve)—color sorting. Then, four commercialized Platycodon grandiflorum seed lots were sorted by this process using corresponding parameters in the actual processing equipment. Subsequently, the seed clarity and germination percentage were significantly improved, and the seed quality qualification rate was increased from 25% to 75%. In summary, by using machine vision technology to quickly extract the material characteristics of the seeds, combined with correlation analysis, probability density distribution plots, single feature selection, and combination sorting comparisons, the appropriate processing process and corresponding sorting parameters for a specific seed lot can be determined, thus maximizing the seed quality.