High-precision identification of highly similar Pinelliae Rhizoma and adulterated Rhizoma pinelliae pedatisectae through deep neural networks based on vision transformers.

Autor: Chen R; State Key Laboratory of Southwestern Chinese Medicine Resources, School of Ethnic Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China., Zhang Y; Center for Computational Sciences, College of Physics and Electronic Engineering, Sichuan Normal University, Chengdu, China., Song WJ; State Key Laboratory of Southwestern Chinese Medicine Resources, School of Ethnic Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China., Zhao TT; State Key Laboratory of Southwestern Chinese Medicine Resources, School of Ethnic Medicine, Chengdu University of Traditional Chinese Medicine, Chengdu, China., Wang JN; Center for Computational Sciences, College of Physics and Electronic Engineering, Sichuan Normal University, Chengdu, China., Zhao YH; Center for Computational Sciences, College of Physics and Electronic Engineering, Sichuan Normal University, Chengdu, China.
Jazyk: angličtina
Zdroj: Journal of food science [J Food Sci] 2024 Nov; Vol. 89 (11), pp. 7372-7379. Date of Electronic Publication: 2024 Oct 09.
DOI: 10.1111/1750-3841.17440
Abstrakt: Pinelliae Rhizoma is a key ingredient in botanical supplements and is often adulterated by Rhizoma Pinelliae Pedatisectae, which is similar in appearance but less expensive. Accurate identification of these materials is crucial for both scientific and commercial purposes. Traditional morphological identification relies heavily on expert experience and is subjective, while chemical analysis and molecular biological identification are typically time consuming and labor intensive. This study aims to employ a simpler, faster, and non-invasive image recognition technique to distinguish between these two highly similar plant materials. In the realm of image recognition, we aimed to utilize the vision transformer (ViT) algorithm, a cutting-edge image recognition technology, to differentiate these materials. All samples were verified using DNA molecular identification before image analysis. The result demonstrates that the ViT algorithm achieves a classification accuracy exceeding 94%, significantly outperforming the convolutional neural network model's 60%-70% accuracy. This highlights the efficiency of this technology in identifying plant materials with similar appearances. This study marks the pioneer work of the ViT algorithm to such a challenging task, showcasing its potential for precise botanical material identification and setting the stage for future advancements in the field.
(© 2024 Institute of Food Technologists.)
Databáze: MEDLINE
Externí odkaz: