Full
Outline
In response to the problem of large computation and slow detection speed of the current pedestrian detection algorithm in underground mines, a lightweight underground pedestrian detection model named MBE-YOLO was proposed based on YOLOv8n. Firstly, a lighter neck structure was designed based on the weighted bidirectional feature pyramid network, with the goal of efficiently fusing multi-scale information. Secondly, the detection head of the original network was redesigned using shared convolution, and the phantom convolution was introduced to obtain rich and effective feature representations, thus solving the problem of the huge number of parameters in the original detection head. Finally, considering the impact of lightweight model on accuracy, a lightweight hybrid local channel attention mechanism was added, which adaptively adjusts the importance of different regions at the global and local levels to improve accuracy. Experimental results show that compared with the benchmark model YOLOv8n, the mAP@0.5:0.95 of MBE-YOLO model is improved by 0.2 percentage points, the computational cost is reduced by 44%, the FPS is increased by 4.3 frames/s, the number of parameters is reduced by 46%, and the model size is reduced by 43%. Compared with the five mainstream models, this model has significant advantages in detection accuracy and lightweight, making it more suitable for scenarios that require real-time detection.
55
27
| 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
