激光雷达空洞率特征与深度学习在复杂煤矸分选中的应用研究

耿皓宇; 乔鑫; 于岳辰; 杨扬; 薛德强; 邢冀川

您当前的位置：

首页 >

文章列表页 >

激光雷达空洞率特征与深度学习在复杂煤矸分选中的应用研究

光学传感与测量 | 更新时间：2026-07-23

- 激光雷达空洞率特征与深度学习在复杂煤矸分选中的应用研究
- Research on the application of lidar void rate features and deep learning in complex coal–gangue separation
- 光学技术 2026年52卷第4期页码：443-451
- 作者机构：
  
  1.山东科技大学机械电子工程学院，山东青岛 266590
  2.中国人民解放军93160部队，北京 100071
  3.西安现代控制技术研究所，陕西西安 710065
  4.内蒙古上海庙矿业有限责任公司榆树井煤矿，内蒙古鄂尔多斯 016200
  5.北京理工大学光电学院，北京 100081
- 作者简介：
  
  耿皓宇（2002— ），男，硕士研究生，从事煤矸识别研究。m15062271505@163.com
  邢冀川（1974— ），男，讲师，博士，从事结合人工智能的光电测试检测研究。michaelhsing@bit.edu.cn
- 基金信息：
  
  国家自然科学基金(52304143;5574198);山东省高等学校青年创新团队发展计划（科技类）(2024KJH061);2025年山东省青年科技人才托举工程项目(SDAST2025QTA055)
- DOI：
  中图分类号： TP23;TP391.4
- 收稿：2026-03-14，
  
  修回：2026-04-13，
  
  录用：2026-05-09，
  
  网络首发：2026-07-23，
  
  纸质出版：2026-07-20
- 稿件说明：
移动端阅览
耿皓宇,乔鑫,于岳辰等.激光雷达空洞率特征与深度学习在复杂煤矸分选中的应用研究[J].光学技术,2026,52(04):443-451.

GENG Haoyu,QIAO Xin,YU Yuechen,et al.Research on the application of lidar void rate features and deep learning in complex coal–gangue separation[J].OPTICAL TECHNIQUE,2026,52(04):443-451.
耿皓宇,乔鑫,于岳辰等.激光雷达空洞率特征与深度学习在复杂煤矸分选中的应用研究[J].光学技术,2026,52(04):443-451. DOI：

GENG Haoyu,QIAO Xin,YU Yuechen,et al.Research on the application of lidar void rate features and deep learning in complex coal–gangue separation[J].OPTICAL TECHNIQUE,2026,52(04):443-451. DOI：

摘要

为提升煤矸识别的可靠性，文章重点研究了激光雷达图像中的空洞率在深度学习分类任务中的作用。针对煤与矸石在采集中出现的激光回波缺失现象，分析了其形成机制及采集角度、物料湿度等因素对空洞率分布的影响。通过对大量激光雷达点云与伪彩图像的对比实验，验证了空洞率在表征煤、矸材质差异方面的稳定性及其对深度网络判别能力的增强效果。进一步利用 DenseNet 模型进行训练后，引入空洞率相关特征的煤矸石识别方法在复杂工况下表现出更高的分类精度与鲁棒性，整体准确率达到94.56%，相比基于地面激光扫描仪强度信息和 K-means与 SVM 算法对泥岩与砂岩分类识别的精度提高了4%以上。研究证明空洞率是一项具有工程应用潜力的关键识别特征，为激光雷达在矿物智能分选中的应用提供了新的思路。

Abstract

Coal–gangue identification is essential for ensuring the efficiency and stability of coal production

yet conventional sensing methods often show limited robustness in complex mining environments. This study aims to enhance identification performance by exploring void rate features in LiDAR data and evaluating their effectiveness in deep learning–based classification. First

the formation mechanism of laser-echo loss in LiDAR acquisition is analyzed

and the effects of scanning angle and material moisture on the spatial distribution of void rate characteristics are investigated. Large-scale LiDAR point-cloud data are then converted into pseudo-color images

from which void rate features are extracted through preprocessing and segmentation. These features are incorporated into a DenseNet-based classification framework. Comparative experiments are conducted to evaluate the contribution of void rate information against conventional single-channel inputs. The results demonstrate that void rate features provide a stable and discriminative representation of material differences between coal and gangue. Incorporating void rate information significantly improves classification accuracy and robustness under varying operating conditions. The findings confirm that void rate is an effective engineering feature with strong practical value and offer new insights into the application of LiDAR technology for intelligent mineral separation in complex industrial environments.

关键词

Keywords

references

Han S , Chen H , Long R , et al . Peak coal in china: A literature review [J ] . Resources, Conservation and Recycling , 2018 , 129 : 293 — 306 .

Leonard R , Zulfikar R , Stansbury R . Coal mining and lung disease in the 21st century [J ] . Current Opinion in Pulmonary Medicine , 2020 , 26 （ 2 ）: 135 — 141 .

Zhao G , Chang F , Chen J , et al . Research and prospect of underground intelligent coal gangue sorting technology: A review [J ] . Minerals Engineering , 2024 , 215 : 108818 .

Jiang Z , He K , Zhang D . A review of intelligent coal gangue separation technology and equipment development [J ] . International Journal of Coal Preparation and Utilization , 2024 , 44 （ 9 ）: 1308 — 1324 .

Liu Q , Li J , Li Y , et al . Recognition methods for coal and coal gangue based on deep learning [J ] . Ieee Access , 2021 , 9 : 77599 — 77610 .

Li D , Wang G , Zhang Y , et al . Coal gangue detection and recognition algorithm based on deformable convolution YOLOv3 [J ] . IET Image Processing , 2022 , 16 （ 1 ）: 134 — 144 .

Yan P , Sun Q , Yin N , et al . Detection of coal and gangue based on improved YOLOv5. 1 which embedded scSE module [J ] . Measurement , 2022 , 188 : 110530 .

Lai W , Hu F , Kong X , et al . The study of coal gangue segmentation for location and shape predicts based on multispectral and improved Mask R-CNN [J ] . Powder Technology , 2022 , 407 : 117655 .

Wang S , Zhu J , Li Z , et al . Coal gangue target detection based on improved YOLOv5s [J ] . Applied Sciences , 2023 , 13 （ 20 ）: 11220 .

Xue G , Li S , Hou P , et al . Research on lightweight Yolo coal gangue detection algorithm based on resnet18 backbone feature network [J ] . Internet of Things , 2023 , 22 : 100762 .

Sui Y , Zhang L , Sun Z , et al . Research on coal and gangue recognition based on the improved YOLOv7-tiny target detection algorithm [J ] . Sensors , 2024 , 24 （ 2 ）: 456 .

Yuan X , Fu Z , Zhang B , et al . Research on lightweight algorithm for gangue detection based on improved Yolov5 [J ] . Scientific Reports , 2024 , 14 （ 1 ）: 6707 .

Cao Z , Li Z , Fang L , et al . Lightweight coal and gangue detection algorithm based on improved Yolov7-tiny [J ] . International Journal of Coal Preparation and Utilization , 2024 , 44 （ 11 ）: 1773 — 1792 .

Zhou M , Lai W . Coal gangue recognition based on spectral imaging combined with XGBoost [J ] . PloS One , 2023 , 18 （ 1 ）: e0279955 .

Yan P , Kan X , Zhang H , et al . Target recognition of coal and gangue based on improved YOLOv5s and spectral technology [J ] . Sensors , 2023 , 23 （ 10 ）: 4911 .

Zhao L , Han L , Zhang H , et al . Study on recognition of coal and gangue based on multimode feature and image fusion [J ] . Plos One , 2023 , 18 （ 2 ）: e0281397 .

Hao W , Jiang H , Song Q , et al . A multi modal fusion coal gangue recognition method based on IBWO-CNN-LSTM [J ] . Scientific Reports , 2024 , 14 （ 1 ）: 30396 .

Xing J , Zhao Z , Wang Y , et al . Coal and gangue identification method based on the intensity image of lidar and DenseNet [J ] . Applied Optics , 2021 , 60 （ 22 ）: 6566 — 6572 .

Sheng J , Liu Z , Zeng Q , et al . Sand bodies identification in reservoir outcrops using intensity from terrestrial laser scanner [C ] // SEG International Exposition and Annual Meeting . SEG , 2016 : SEG-2016- 13772603 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于柱状物点云语义特征的相机激光联合标定算法

一种激光雷达轴棱锥光学空间滤波器

基于DOAS技术检测痕量气体的研究综述

基于双光楔的长波像方扫描光学系统设计

融合深度空间校正与几何锚点的单目视觉行道树参数测量