English
高级检索

多尺度特征融合的露天矿区道路负障碍检测

阮顺领, 李少博, 卢才武, 顾清华

阮顺领, 李少博, 卢才武, 等. 多尺度特征融合的露天矿区道路负障碍检测[J]. 煤炭学报, 2021, 46(S2): 1170-1179.
引用本文: 阮顺领, 李少博, 卢才武, 等. 多尺度特征融合的露天矿区道路负障碍检测[J]. 煤炭学报, 2021, 46(S2): 1170-1179.
shu
RUAN Shunling, LI Shaobo, LU Caiwu, et al. Road negative obstacle detection in open⁃pit mines based on multi scale feature fusion[J]. Journal of China Coal Society, 2021, 46(S2): 1170-1179.
Citation: RUAN Shunling, LI Shaobo, LU Caiwu, et al. Road negative obstacle detection in open⁃pit mines based on multi scale feature fusion[J]. Journal of China Coal Society, 2021, 46(S2): 1170-1179.
shu

多尺度特征融合的露天矿区道路负障碍检测

  • 西安建筑科技大学 资源工程学院;西安市智慧工业感知计算与决策重点实验室

Road negative obstacle detection in open⁃pit mines based on multi scale feature fusion

  • School of Resource Engineering，Xi’ an University of Architecture and Technology; Xi’ an Key Laboratory of Intelligent Industry Perception Computing and Decision Making，Xi’ an University of Architecture and Technology

摘要

    摘要
  • 摘要: 露天矿区非结构化道路存在的坑洼、道路塌陷等负向障碍，易导致车辆侧翻或陷车，近年来 矿用卡车自动驾驶的兴起，使得负向障碍检测变得至关重要。 对露天矿区道路负障碍特征进行深 入分析，构建了基于机器视觉的轻量化目标检测模型。 首先通过现场采样及标注建立露天矿区负 向障碍数据集并将其输入到目标检测模型;其次对图像进行归一化处理并使用 MobileNetv3 网络对 图像进行压缩激活，在获得输出特征后进行连续上采样和特征金字塔堆叠，完成多尺度特征提取; 最后对多尺度特征进行分类和边界框回归，达到负向障碍检测的目的。 模型在特征金字塔模块中 引入深度可分离卷积方法，降低网络特征提取和融合的计算量;通过对损失函数和学习率动态优化 调整，提高负障碍目标检测精度;在负障碍检测后处理阶段，提出非极大抑制优化算法，改善负障碍 被遮挡和检测框定位精度不高的问题。 实验结果表明，研究方法能有效识别复杂背景下露天矿区 非结构化道路的负障碍，对矿区道路负障碍检测精度、召回率、mAP 达到 98.86%，89.58%和 92.59%，实时检测速度达到 47. 3 fps。 对比主流的目标检测网络，如:Yolov3，RetianNet，SSD，Faster RCNN 等也有着更好的综合性能。 通过对非极大抑制优化算法的量化分析，相较于传统算法，改进 算法可有效改善检测框的定位精度，同时具有良好的适应性。
    Abstract: The existence of negative obstacles such as potholes and road cave⁃ins on unstructured roads in open pit mines can easily lead to vehicle rollover or trapping，and the development of autonomous mining truck driving in re⁃ cent years has made negative obstacle detection crucial. In this study，the characteristics of road negative obstacles in open⁃pit mining area are comprehensively analyzed，and a lightweight object detection model based on machine vision is constructed. Firstly，the negative obstacle data set of open⁃pit mining area is constructed by field sampling and input into the target detection model. Secondly，the image is normalized and activated by mobileNetv3 network. After obtai⁃ ning the output features，the continuous up⁃sampling and feature pyramid stacking are performed to complete multi⁃ scale feature extraction. Finally，the multi⁃scale features are classified and the bounding box regression is performed to achieve negative obstacle detection. The model introduces a depth⁃separable convolution method in the feature pyramid module to reduce the computational effort of network feature extraction and fusion. By adjusting the loss function and learning rate dynamically and optimally，the accuracy of negative obstacle target detection is improved. In the post⁃ processing stage of negative obstacle detection，a non⁃maximum suppression optimization algorithm is proposed in the post⁃processing stage of negative obstacle detection to solve the problem that the negative obstacles are obscured and the detection frame is not localized with high accuracy. The experimental results show that the research meth⁃ od can effectively identify negative obstacles on unstructured roads in open⁃pit mining areas in complex backgrounds. The mine road negative barrier detection precision，recall and mAP reach98. 86%，89. 58% and 92. 59% with a real time detection speed of 47.3 fps. It also has a better overall performance than that of mainstream target detec⁃ tion networks such as Yolov3，RetianNet，SSD，Faster RCNN etc. A quantitative analysis of the non⁃maximal suppres⁃ sion optimization algorithm has also shown that the improved algorithm is effective in improving the positioning accura⁃ cy of the detection boxes compared to the traditional algorithm and is also very adaptable.
    • HTML全文
    • 参考文献(0)
    • 相关文章
    • 施引文献(18)

  • 期刊类型引用(12)

    1. 陈良. 基于RT-DETR的露天矿区路面障碍物检测. 黄金科学技术. 2025(01): 202-213 . 百度学术
    2. 徐勇超，佟佳庆，黄磊. 边缘计算技术在露天矿人车防撞实时处理中的应用研究. 信息与电脑. 2025(06): 127-129 . 百度学术
    3. 顾清华，周琼，王丹. 基于改进YOLOv8的露天矿区行车障碍物检测. 黄金科学技术. 2024(02): 345-355 . 百度学术
    4. 岳中文，金庆雨，潘杉，颜文婧，覃逸峰，陈震. 基于深度学习的轻量化炮孔智能检测方法. 煤炭学报. 2024(05): 2247-2256 . 本站查看
    5. 阮顺领，鄢盛钰，顾清华，卢才武. 基于多特征融合的露天矿区道路负障碍检测. 煤炭学报. 2024(05): 2561-2572 . 本站查看
    6. 韩广涛，张明路，高春艳，吕晓玲. 室外广域环境下移动机器人地形障碍检测及分析方法. 科学技术与工程. 2024(19): 8150-8157 . 百度学术
    7. 阮顺领，张回国，顾清华，卢才武，刘迪，毛晶. 基于双目视觉的露天矿无人车前障碍检测研究. 煤炭学报. 2024(S2): 1285-1294 . 本站查看
    8. 杨春雨，陈佳怡，张鑫，唐超权. 矿用卡车路面自适应模型预测轨迹跟踪控制. 控制理论与应用. 2023(06): 1061-1068 . 百度学术
    9. 毕松，王远航，李嘉伟，刘蕾. 户外种植园行间可通行区域识别方法研究. 计算机仿真. 2023(07): 423-428 . 百度学术
    10. 顾清华，杜艺凡，李萍丰，王丹. 基于加权双向特征融合的矿区道路落石检测. 黄金科学技术. 2023(06): 953-963 . 百度学术
    11. 阮顺领，董莉娟，卢才武，顾清华. 基于RCR＿YOLOv4的矿井巷道红外障碍检测研究. 黄金科学技术. 2022(04): 603-611 . 百度学术
    12. 詹云峰，黄志斌，付波，王小龙. 基于视觉的水面背景下目标检测与跟踪算法. 科学技术与工程. 2022(33): 14809-14819 . 百度学术

    其他类型引用(6)

    • 资源附件(0)
计量
  • 文章访问数:  110
  • HTML全文浏览量:  8
  • PDF下载量:  210
  • 被引次数: 18
出版历程
  • 网络出版日期:  2023-04-09

目录

摘要
HTML全文
    参考文献(0)
    相关文章
    施引文献(18)
    资源附件(0)

    /

    返回文章
    返回

    相关链接

    邮件订阅 RSS

    网站版权 © 《煤炭学报》编辑部 京ICP备05086979号-14

    地址：北京市朝阳区和平街13区煤炭大厦

    邮编：100013

    编辑部电话：（010）87986411

    本系统由 北京仁和汇智信息技术有限公司 开发  百度统计

    x 关闭 永久关闭