Lightweight deep learning algorithm for real-time wheat flour quality detection via NIR spectroscopy

Yu Yang; Rumeng Sun; Hongyan Li; Yao Qin; Qinghui Zhang; Pengtao Lv; Quan Pan

doi:10.1016/j.saa.2024.125653

Lightweight deep learning algorithm for real-time wheat flour quality detection via NIR spectroscopy

Spectrochim Acta A Mol Biomol Spectrosc. 2024 Dec 22:330:125653. doi: 10.1016/j.saa.2024.125653. Online ahead of print.

Authors

Yu Yang¹, Rumeng Sun², Hongyan Li¹, Yao Qin¹, Qinghui Zhang¹, Pengtao Lv¹, Quan Pan³

Affiliations

¹ Key Laboratory of Grain Information Processing and Control (Henan University of Technology), Ministry of Education, Zhengzhou 450001, China; Henan Key Laboratory of Grain Storage Information Intelligent Perception and Decision Making, Henan University of Technology, Zhengzhou 450001, China; Henan Grain Big Data Analysis and Application Engineering Research Center (Henan University of Technology), Zhengzhou 450001, China; College of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, China.
² College of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, China.
³ College of Information Science and Engineering, Henan University of Technology, Zhengzhou 450001, China; School of Automation, Northwestern Polytechnical University, Xi'an, China. Electronic address: [email protected].

PMID: 39733712
DOI: 10.1016/j.saa.2024.125653

Abstract

Wheat flour quality, determined by factors such as protein and moisture content, is crucial in food production. Traditional methods for analyzing these parameters, though precise, are time-consuming and impractical for large-scale operations. This study presents a lightweight convolutional neural network designed for real-time wheat flour quality monitoring using near-infrared spectroscopy. The model incorporates Ghost bottlenecks, external attention modules, and the Kolmogorov-Arnold network to enhance feature extraction and improve prediction accuracy. Testing results demonstrate high predictive performance with R² values of 0.9653 (RMSE: 0.2886 g/100 g, RPD: 5.8981) for protein and 0.9683 (RMSE: 0.3061 g/100 g, RPD: 5.1046) for moisture content. The model's robustness across diverse samples and its suitability for online applications make it a promising tool for efficient and non-destructive quality control in the food industry.

Keywords: Lightweight convolutional neural network; Near-infrared spectroscopy; Non-destructive food quality control; Online monitoring.