Full
Outline
Aiming at the problem that the convolutional neural network applied in the existing copper ore grade classification lacked the relationship between the construction and the induction of long-distance features, a copper ore grade classification method based on the integrated model of Swin Transformer and EfficientNet was proposed. This method made full use of the inductive ability of Swin Transformer V2-t architecture for long-distance feature relations and the advantage of EfficientNet V2-s in capturing subtle local features. By adding a linear layer to integrate the output results of the two models, and dynamically adjusting the weight of the linear layer according to the output of a single model itself, the mapping relationship was optimized, and the classification performance was significantly improved. The experimental results show that the accuracy rate of the fusion model in the classification task is 92.891%, the accuracy rate is 93.095%, and the recall rate is 92.654%. Compared with the single model before integration, the integrated comprehensive model has improved the classification accuracy by 1.3 percentage points, the accuracy rate by 1.9 percentage points and 2.186 percentage points respectively, and the recall rate by 0.474 percentage points and 0.237 percentage points respectively.
47
21
| 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
