Improved YOLOv11 Target Detection Model for Complex Coal Mine Environments

NAVIGATE

Table of Contents

STATISTICS

Viewed53

Downloads46

Improved YOLOv11 Target Detection Model for Complex Coal Mine Environments

PDF下载 (46)

[1]ZHANG Jianhui,CAI Xiaohang,et al.Improved YOLOv11 Target Detection Model for Complex Coal Mine Environments[J].Journal of Zhengzhou University (Engineering Science),2027,48(XX):1-8.[doi:10.13705/j.issn.1671-6833.2026.04.009]

Copy

Journal of Zhengzhou University (Engineering Science)[ISSN 1671-6833/CN 41-1339/T] Volume: 48 Number of periods: 2027 XX Page number: 1-8 Column: Public date: 2027-12-10

Title:: Improved YOLOv11 Target Detection Model for Complex Coal Mine Environments

Author(s):: ZHANG Jianhui ^{1, 2} , CAI Xiaohang ¹ , WANG Ruimin ³ , ZENG Junjie ¹ , LUO Xudong¹; 1. School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450002, China; 2. Songshan Laboratory, Zhengzhou 450002, China; 3. School of Computer Science and Artificial Intelligence, Zhengzhou University, Zhengzhou 450001, China

Keywords:: YOLOv11; target detection; dynamic inception mixer convolution; multi-attention mechanism; efficient upsampling

CLC:: TP391 ；TD76

DOI:: 10.13705/j.issn.1671-6833.2026.04.009

Abstract:: To address the insufficient target detection accuracy caused by harsh working conditions in coal mine construction environments, such as uneven illumination distribution, severe target occlusion, and dust interference, a target detection model named DME-YOLO was proposed for coal mine in complex environments based on DIM and YOLOv11. In the backbone network of DME-YOLO, a dynamic inception mixer convolution module (DIM) was designed. This module achieved adaptive fusion of multi-scale features through a dynamic weight mechanism, thereby enhancing the model’s capability of feature representation in complex backgrounds. For the detection head, a dynamic multi-attention detection head (DMA-Head) was introduced, which leveraged a multi-scale attention module to strengthen the perception of small targets and targets with weak textures. Additionally, an efficient upsampling convolutional block (EUCB) was embedded into the neck network optimizing the upsampling path by combining bilinear interpolation with depthwise separable convolution. Experimental results demonstrated that DME-YOLO achieved a mAP@50 of 93.7% on the self-constructed mine dataset, representing 3.0 percentage points improvement compared to the original YOLOv11. Its mAP@50-95 reached 66.8%, which was 5.2 percentage points increase relative to the original YOLOv11. When compared with models such as YOLOv9s and YOLOv12, DME-YOLO exhibited faster convergence speed and superior detection accuracy, making it well-suited for safety monitoring in coal mine construction sites.

References:: [1] Yuan Zhi, Jiang Qingyou, Pang Zhenzhong. Application status and development thinking of intelligent mining technology and equipment in coal mines in China[J]. Coal Science and Technology, 2024, 52(9): 189-198. [袁智, 蒋庆友, 庞振忠. 我国煤矿智能化综采开采技术装备应用现状与发展思考[J]. 煤炭科学技术, 2024, 52(9): 189-198.]
[2] Jung D, Choi Y. Systematic review of machine learning applications in mining: Exploration, exploitation, and reclamation[J]. Minerals, 2021, 11(2): 148.
[3] Girshick R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2016: 1440-1448.
[4] He Kaiming, Gkioxari G, Dollár P, et al. Mask R-CNN[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2017: 2980-2998.
[5] Redmon J, Farhadi A. YOLOv3: an incremental improvement[PP/OL]. V1. arXiv(2018-04-08)[2025-09-01]. https://arxiv.org/abs/1804. 02767.
[6] Ultralytics. YOLOv5[EB/OL]. (2020-05-18)[2025-09-01]. https://github.com/ultralytics/yolov5.
[7] Wang C Y, Bochkovskiy A, Liao H M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2023: 7464-7475.
[8] Ultralytics. Ultralytics[EB/OL]. (2023-01-10)[2025-09-01]. https://github.com/ultralytics/ultralytics.
[9] WANG C Y, YEH I H, LIAO H Y M. YOLOv9: learning what you want to learn using programmable gradient information[PP/OL]. V2. arXiv(2024-02-29)[2025-09-01]. https://arxiv.org/abs/2402. 13616.
[10] Liu Wei, Anguelov D, Erhan D, et al. SSD: single shot MultiBox detector[C]//Computer Vision – ECCV 2016. Cham: Springer, 2016: 21-37.
[11] Zhang Zhihao, Tao Lei, Yao Linhu, et al. LDSI-YOLOv8: Real-time detection method for multiple targets in coal mine excavation scenes[J]. IEEE Access, 2024, 12: 132592-132604.
[12] Ramyadevi R, Keerthy S, Catherina J S J, et al. Helmet and equipment detection with worker’s mobility tracker in mining sector using YOLOv8 & LSTM[C]//Proceedings of the 2025 IEEE International Students’ Conference on Electrical, Electronics and Computer Science (SCEECS). Piscataway: IEEE, 2025: 1-6.
[13] Zhang Lei, Sun Zhipeng, Tao Hongjing, et al. Research on mine-personnel helmet detection based on multi-strategy-improved YOLOv11[J]. Sensors, 2025, 25(1): 170.
[14] Shao Xiaoqiang, Liu Shibo, Li Xin, et al. Rep-YOLO: an efficient detection method for mine personnel[J]. Journal of Real-Time Image Processing, 2024, 21(2): 28.
[15] Fu Zhibo, Ling Jierui, Yuan Xinpeng, et al. Yolov8n-FADS: a study for enhancing miners’ helmet detection accuracy in complex underground environments[J]. Sensors, 2024, 24(12): 3767.
[16] Yu Weihao, Si Chenyang, Zhou Pan, et al. MetaFormer baselines for vision[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(2): 896-912.
[17] Yu Ziping, Huang Hongbo, Chen Weijun, et al. YOLO-FaceV2: a scale and occlusion aware face detector[J]. Pattern Recognition, 2024, 155: 110714.
[18] Rahman M M, Munir M, Marculescu R. EMCAD: efficient multi-scale convolutional attention decoding for medical image segmentation[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2024: 11769-11779.
[19] Wang Jiaqi, Chen Kai, Xu Rui, et al. CARAFE: content-aware reassembly of FEatures[C]//Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2019: 3007-3016.
[20] Liu Wenze, Lu Hao, Fu Hongtao, et al. Learning to up-sample by learning to sample[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2024: 6004-6014.
[21] Lu Hao, Liu Wenze, Ye Zixuan, et al. SAPA: similarity-aware point affiliation for feature upsampling[PP/OL]. V2. arXiv(2022-12-27)[2025-10-20]. https://arxiv.org/abs/2209.12866.
[22] Lu Hao, Liu Wenze, Fu Hongtao, et al. FADE: fusing the assets of decoder and encoder for task-agnostic upsampling[PP/OL]. V2. arXiv(2022-12-27)[2025-09-01]. https://arxiv.org/abs/2207.10392.
[23] Xie Weiming, Ma Weifeng, Sun Xiaoyong. An efficient re-parameterization feature pyramid network on YOLOv8[J]. Neurocomputing, 2025, 614: 128775.
[24] Dai Xiyang, Chen Yinpeng, Xiao Bin, et al. Dynamic head: unifying object detection heads with attentions[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2021: 7369-7378.
[25] Zhang Jiarui, Chen Zhihua, Yan Guoxu, et al. Faster and lightweight: an improved YOLOv5 object detector for remote sensing images[J]. Remote Sensing, 2023, 15(20): 4974.
[26] Gao Lin, Yu Pengwei, Dong Hongjuan, et al. Multi-scale fusion lightweight target detection method for coal and gangue based on EMBS-YOLOv8s[J]. Sensors, 2025, 25(6): 1734.

Similar References:

Memo

Last Update: 2026-03-13