Detection of small objects in open-pit coal mine stopes using UAV remote sensing images

被引：0
作者：
Liu M. [1 ]
Che B. [2 ]
Dong H. [3 ]
Zhu R. [4 ]
机构：
[1] College of Geological Engineering and Geomatics, Chang’an University, Xi’an
[2] Guangzhou Institute of Technology, Xidian University, Guangzhou
[3] CCTEG Xi’an Research Institute (Group) Co., Ltd., Xi’an
[4] Department of Economics, Queen’s University, Kingston
来源：
Meitiandizhi Yu Kantan/Coal Geology and Exploration | 2023年 / 51卷 / 11期
关键词：
attention mechanism; Normalized Wasserstein Distance (NWD) metric; open-pit coal mine; remote sensing images of mining area; small object detection; Unmanned aerial vehicle (UAV); YOLOv7;
D O I：
10.12363/issn.1001-1986.23.04.0186
中图分类号：
学科分类号：
摘要：
Open-pit coal mine stopes witness frequent vehicle accidents due to their complex terrains. Therefore, accurately positioning vehicles in the stopes is critical to safe coal mining. To overcome the inaccurate positioning of small targets in the mining areas using Unmanned aerial vehicle (UAV) remote sensing images, this study proposed an improved YOLOv7 model for object detection. First, to expedite the reasoning of the YOLOv7 network, the ELAN module in the original YOLOv7 model was improved by introducing partial convolution. Based on this, the eSE channel attention mechanism was combined to form the PConv-eSE convolution attention module. The purpose is to enhance the network’s ability to extract the features of small targets and reduce the impact of background information. Finally, the loss function of the Normalized Wasserstein Distance (NWD) metric was used to further optimize the network and improve the accuracy. The improved YOLOv7 model was verified through experiments using a vehicle dataset of a mining area stope. The results show that the Pmav value of the improved YOLOv7 model reached 94.5%, which was 7.2% higher than the original model, thus effectively overcoming the missing of small targets from remote sensing images in the detection using the original network. This study will provide a theoretical basis for the positioning of small objects in an open-pit coal mining area using UAVs. © 2023 The Author(s).
引用
收藏
页码：132 / 140
页数：8
相关论文
共 23 条
  • [1] (2020)
  • [2] WANG Kun, Peng YANG, Et al., Current status and development trend of UAV remote sensing applications in the mining industry[J], Chinese Journal of Engineering, 42, 9, pp. 1085-1095, (2020)
  • [3] CHENG Jian, YE Liang, GUO Yinan, Et al., An aerial image detection method of ground crack in goaf based on deformable convolutional network with hybrid domain attention[J], Journal of China Coal Society, 45, Sup.2, pp. 993-1002, (2020)
  • [4] QIN Xiaohui, HUANG Qidong, CHANG Dengxiang, Et al., Object detection method in open−pit mine based on improved YOLOv5[J], Journal of Hunan University (Natural Sciences), 50, 2, pp. 23-30, (2023)
  • [5] XUE Yali, SUN Yu, MA Hanrong, Lightweight small target detection in aerial remote sensing image[J], Electronics Optics & Control, 29, 6, (2022)
  • [6] Zhiwei WEI, DUAN Chengzhen, SONG Xinghao, Et al., Amrnet：Chip augmentation in aerial image object detection[J], (2020)
  • [7] Ruihong YIN, Wei ZHAO, Xudong FAN, Et al., AF–SSD：An accurate and fast single shot detector for high spatial remote sensing imagery[J], Sensors, 20, 22, (2020)
  • [8] ZAND M，, ETEMAD A，, GREENSPAN M., Oriented bounding boxes for small and freely rotated objects[J], IEEE Transactions on Geoscience and Remote Sensing, 60, (2020)
  • [9] KIM M，, JEONG J, KIM S., ECAP–YOLO：Efficient channel attention pyramid YOLO for small object detection in aerial image[J], Remote Sensing, 13, 23, (2021)
  • [10] REDMON J, FARHADI A., YOLOv3：An incremental improvement[J], (2018)
123