Full
Outline
A multi-dimensional expansion lightweight model MD-YOLO based on YOLOv9 model was proposed to address the issues of low detection accuracy, missed detection and false detection of personnel wearing safety helmets in complex mining environments. To solve the problems of low image clarity and dust obstruction, the original feature extraction module was improved by introducing a multi-dimensional expansion module, which enabled the model to obtain a larger receptive field when detecting safety helmets and enhanced the ability to capture global features. To address the missed detection of small and density targets, a dynamic detection head (PD Dyhead) with attention mechanism was constructed, which enhanced the feature extraction ability of the model from three aspects of scale, space and task perception. The use of exponential moving average sample weighted loss function (EMA SledeLoss) improved the shortcomings of YOLOv9 model in dealing with sample imbalance problems. The experimental results show that compared with the YOLOv9 model, the detection accuracy of the MD-YOLO model reaches 92.5%, with an increase of 2.8 percentage points. The parameter quantity decreases by 26.6%, and has a 27.0% increase in detection speed. The MD-YOLO model can effectively monitor the wearing of safety helmets in various complex environments, with the best comprehensive performance.
83
42
| 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
