Application of artificial intelligence in the rapid determination of moisture content in medicine food homology substances.

 0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Tingting Feng, Shouyao Huang, Chang Li, Lian Li, Boran Lin, Lei Nie, Cheng Peng, Wentian Wang, Aoli Wu, Chunguo Yang, Hengchang Zang, Mengyu Zhang, Shudi Zhang

Ngôn ngữ: eng

Ký hiệu phân loại: 364.04 Special topics of criminology

Thông tin xuất bản: England : Food chemistry , 2025

Mô tả vật lý:

Bộ sưu tập: NCBI

ID: 726803

Moisture content is crucial in quality testing of medicine food homology substances. This study aimed to present a new modeling method for moisture content based on near-infrared spectroscopy. When comparing three methods of partial least squares regression, support vector regression and convolutional neural network (CNN) to build moisture content prediction models of three different substances, it was found that the accuracy was affected by systematic error and was low. Thus, this study integrated moisture characteristic bands data of three substances and established a universal model. The optimal prediction model was SSA-CNN-BiLSTM. The RMSEP value of test set were 0.3568 %, 0.2057 % and 0.0029 %, and RPD value were 10.26, 2.30 and 5.60, respectively. The innovation of this study was that the above method improved modeling accuracy and efficiency. It eliminated systematic errors when each substance was modeled individually, simplified the modeling process, and expanded the scope of model application.
Tạo bộ sưu tập với mã QR

THƯ VIỆN - TRƯỜNG ĐẠI HỌC CÔNG NGHỆ TP.HCM

ĐT: (028) 36225755 | Email: tt.thuvien@hutech.edu.vn

Copyright @2024 THƯ VIỆN HUTECH