An online automatic sorting system for defective Ginseng Radix et Rhizoma Rubra using deep learning

Xue, Qilong; Miao, Peiqi; Miao, Kunhong; Yu, Yang; Li, Zheng

doi:10.1016/j.chmed.2023.01.001

Cited by 2 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In view of the problem of detecting internal defects, X-ray based non-destructive testing technology has been combined with deep learning. In this regard, this paper provides a strong basis for the subsequent development of classi cation models for Ephedra seed quality detection (Xue et al, 2023).…”

Section: Discussionmentioning

confidence: 99%

Detection and identification of Ephedra Herba seed pests based on X-rays and DNA barcodes

xiao,

Guo,

Zhang

et al. 2024

Preprint

View full text Add to dashboard Cite

As an important natural medicinal resource for humans, Ephedra sinica Stapf often suffers from various biological stresses during its growth process. One of the pests that pose a threat to Ephedra is the Ephedra seed pest. Its primary mode of damage is laying eggs inside the Ephedra seeds, where they develop and eventually emerge as adult wasps, causing significant damage to the seeds. This study aimed to investigate the effectiveness of X-ray for Ephedra seed pest detection, the impact on the use value of Ephedra seeds infested by pests, and the further confirmation of pest species information. The mature Ephedra seeds from the Inner Mongolia Autonomous Region were taken as the teste object. The results showed that the germination rates of three different batches of seeds were 46%, 40%, and 38%, while the seedling emergence rates were only 36%, 30%, and 32%, respectively, significantly lower than that of control healthy seeds 66% and 52%. The phylogenetic tree Neighbor-Joining (NJ) was constructed by extracting the COI sequences of the pest samples, and was identified as a new species of Eurytoma genus based on morphological characteristics. X-ray detection is a commonly used, non-invasive method. Based on non-destructive X-ray testing, the quality of Ephedra seeds could be classified into three types: healthy plump seeds, seeds infested by pests, and underdeveloped hollow seeds. In conclusion, the combined use of X-ray and DNA barcoding methods can achieve rapid and accurate detection and identification of E. sinica seed pests, which is of great significance for the management and control of Ephedra seed production.

show abstract

Section: Discussionmentioning

confidence: 99%