Machine learning-based non-destructive terahertz detection of seed quality in peanut

• Published in: Food Chemistry: X Vol. 23; p. 101675
• Main Authors: Jiang, Weibin, Wang, Jun, Lin, Ruiquan, Chen, Riqing, Chen, Wencheng, Xie, Xin, Hsiung, Kan-Lin, Chen, Hsin-Yu
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 30-10-2024; Elsevier
• Subjects: Aflatoxin; Machine learning; Peanut; Quality identification; Terahertz imaging technology; Peanut; Quality identification; Terahertz imaging technology; Aflatoxin; Machine learning
• ISSN: 2590-1575; 2590-1575
• DOI: 10.1016/j.fochx.2024.101675

Rapid identification of peanut seed quality is crucial for public health. In this study, we present a terahertz wave imaging system using a convolutional neural network (CNN) machine learning approach. Terahertz waves are capable of penetrating the seed shell to identify the quality of peanuts without causing any damage to the seeds. The specificity of seed quality on terahertz wave images is investigated, and the image characteristics of five different qualities are summarized. Terahertz wave images are digitized and used for training and testing of convolutional neural networks, resulting in a high model accuracy of 98.7% in quality identification. The trained THz-CNNs system can accurately identify standard, mildewed, defective, dried and germinated seeds, with an average detection time of 2.2 s. This process does not require any sample preparation steps such as concentration or culture. Our method swiftly and accurately assesses shelled seed quality non-destructively.
•Terahertz imaging with CNN achieves 98.7% accuracy in identifying 5 peanut seed quality types.•Terahertz imaging reveals unique moisture patterns for different peanut seed qualities.•Machine learning automates seed quality assessment without complex sample preparation.•Rapid, non-destructive seed quality detection potential for entire supply chain.