基于FOFE-YOLO的输煤传送带异物检测

doi:10.19286/j.cnki.cci.2026.04.014

摘要
图/表
参考文献(14)
相关文章 (0)

全文: PDF (0 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要

为了解决目前检测模型对输煤传送带中存在异物、大块煤炭的漏检、检测精度低等问题，提出一种基于FOFE-YOLO的输煤传送带异物检测方法。以YOLOv12检测框架为基准，在A2C2f模块中引入ELA注意力机制，增强对异常目标的特征提取效果；将模型中的部分下采样卷积替换为ADown结构，结合卷积与池化进行融合采样，降低因下采样导致的信息丢失；在C3k2模块中融入DConv卷积，借助动态特征捕获机制，提升模型对全局特征的推断性能，从而加强模型对异物及大块煤炭的识别效果。实验结果表明，改进后模型的mAP0.5比基线模型提高了2%，达到了90.6%，适用于实时煤矿安全防护监测。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	杨磊

关键词 ：目标检测, 输煤传送带, 异物检测, YOLOv12

Abstract：

In order to solve the problems of foreign matter in coal conveyor belt, missed detection of large coal and low detection accuracy in the current detection model, a foreign matter detection method of coal conveyor belt based on FOFE-YOLO is proposed. Based on the YOLOv12 detection framework, the ELA attention mechanism is introduced into the A2C2f module to enhance the feature extraction effect of abnormal targets. The partial downsampling convolution in the model is replaced by the ADown structure, and the fusion sampling is combined with convolution and pooling to reduce the information loss caused by downsampling. The DConv convolution is integrated into the C3k2 module, and the dynamic feature capture mechanism is used to improve the inference performance of the model for global features, thereby enhancing the recognition effect of the model on foreign bodies and large coal. The experimental results show that the mAP0.5 of the improved model is 2 % higher than that of the baseline model, reaching 90.6 %, which is suitable for real-time coal mine safety protection monitoring

Key words： target detection coal conveyor belt foreign body detection YOLOv12

基金资助:

冀中能源2025年度科技创新项目（ 2025JZ022 ）

作者简介: 杨磊（ 2001— ），男，河北邢台人。

引用本文:

杨磊. 基于FOFE-YOLO的输煤传送带异物检测[J]. 煤炭与化工, 2026, 49(4): 71-75，80.. Yang Lei. Foreign body detection of coal conveyor belt based on FOFE-YOLO. CCI, 2026, 49(4): 71-75，80..

链接本文:

http://www.mtyhg.com.cn/CN/10.19286/j.cnki.cci.2026.04.014 或 http://www.mtyhg.com.cn/CN/Y2026/V49/I4/71

[ 1 ]       仵    杰，贺巧巧. 基于改进YOLOv8的煤矿井下传送带异物检测研究[ J ]. 中国科技论文，2025，20（ 8 ）：673 - 683.
[ 2 ]       王    辉，韩    磊. 基于改进YOLOv11的煤矿传送带异物检测方法[ J ]. 电子设计工程，2025，33（ 14 ）：66 - 70.
[ 3 ]       Girshick R. Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2015: 1 440 - 1 448.
[ 4 ]       Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox det-
ector[ C ]//European conference on computer vision. Cham: Sprin-
ger International Publishing, 2016: 21 - 37.
[ 5 ]       Wang A, Chen H, Liu L, et al. Yolov10: Real-time end-to-end o-
bject detection[ J ]. Advances in Neural Information Processing Systems, 2024（ 37 ）: 107 984 - 108 011.
[ 6 ]       Khanam R, Hussain M. Yolov11: An overview of the key architec-
tural enhancements[ J ]. arXiv preprint arXiv:2410.17725, 2024.
[ 7 ]       Tian Y, Ye Q, Doermann D. Yolov12: Attention-centric real-time object detectors[ J ]. arXiv preprint arXiv: 2502.12524, 2025.
[ 8 ]       Lei M, Li S, Wu Y, et al. YOLOv13: Real-Time Object Detection with Hypergraph-Enhanced Adaptive Visual Perception[ J ]. arXiv preprint arXiv: 2506.17733, 2025.
[ 9 ]       郭志聪，张    雷. 基于改进YOLOv8的低照度煤矿传送带异物识别算法[ J ]. 电子测量技术，2024，47（ 21 ）：188 - 196.

[ 10 ]     张    杨，程智宇，陈允降，等. 注意力机制增强的输煤传送带异物检测[ J ]. 智能科学与技术学报，2025，7（ 2 ）：268 - 276.
[ 11 ]     林永洪，郑建明，李    照. 基于UD-DETR轻量化网络小目标异物检测算法[ J ]. 信息化研究，2025，51（ 1 ）：71 - 78.
[ 12 ]     Xu W, Wan Y. ELA: Efficient local attention for deep convolutio-
nal neural networks[ J ]. arXiv preprint arXiv:2403.01123, 2024.
[ 13 ]     Wang C Y, Yeh I H, Mark Liao H Y. Yolov9: Learning what you want to learn using programmable gradient information[ C ]//European conference on computer vision. Cham: Springer Nature Switzerland, 2024: 1 - 21.
[ 14 ]     Wang R, Shivanna R, Cheng D, et al. Dcn v2: Improved deep & cross network and practical lessons for web-scale learning to rank systems[ C ]//Proceedings of the web conference 2021. 2021: 1 785 - 1 797.