Full
Outline
To investigate a detection method based on near-infrared hyperspectral imaging combined with decision tree algorithm and minimum noise fraction transform (MNF).
Hyperspectral images of seaweed samples containing impurities were acquired using a GaiaSorter-N17E system (spectral range: 900–1700 nm).Following radiometric calibration, MNF was applied for dimensionality reduction and noise removal to extract feature bands. Subsequently, a decision tree classification model was constructed for impurity identification.
Significant spectral differences were observed among different impurities in the near-infrared region. The first several MNF bands contained the main information and effectively highlighted impurity characteristics. The decision tree model clearly distinguished seaweed from sand, wood, charcoal and plastic rope, demonstrating excellent performance in identifying fine plastic rope. The method achieved a high overall classification accuracy of 92.8% and a fast processing speed of approximately 1.6 s per image without requiring extensive training samples.
This method provides effective technical support for automated impurity detection in seaweed and holds significant potential for application in food industry quality monitoring.
7
1
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
