Improved Rock Detection in Harsh Environments for YOLOv8n

ZOU Shao-fa; WEI Zhen-hua

doi:10.12263/newIT.2025.01.002

您当前的位置：

首页 >

文章列表页 >

Improved Rock Detection in Harsh Environments for YOLOv8n

RESEARCH PAPER | 更新时间：2025-07-04

- Improved Rock Detection in Harsh Environments for YOLOv8n
- New Generation of Information Technology Vol. 8, Issue 1, Pages: 6-10(2025)
- 作者机构：
  
  1.东华理工大学信息工程学院，江西南昌 330013
  2.核技术应用教育部工程研究中心，江西南昌 330013
  3.江西省核地学数据科学与系统工程技术研究中心，江西南昌 330013
- 作者简介：
- 基金信息：
- DOI：10.12263/newIT.2025.01.002
  CLC： TP391.41
- Accepted：18 April 2025，
  
  Published：15 January 2025
- 稿件说明：
移动端阅览
邹绍发, 魏振华. 改进YOLOv8n的恶劣环境下岩石检测方法[J]. 新一代信息技术, 2025, 8(1): 06-10

ZOU Shao-fa, WEI Zhen-hua. Improved Rock Detection in Harsh Environments for YOLOv8n[J]. New Generation of Information Technology, 2025, 8(1): 06-10
邹绍发, 魏振华. 改进YOLOv8n的恶劣环境下岩石检测方法[J]. 新一代信息技术, 2025, 8(1): 06-10 DOI： 10.12263/newIT.2025.01.002.

ZOU Shao-fa, WEI Zhen-hua. Improved Rock Detection in Harsh Environments for YOLOv8n[J]. New Generation of Information Technology, 2025, 8(1): 06-10 DOI： 10.12263/newIT.2025.01.002.

摘要

岩石破碎作为采矿业中获取有价值矿物的关键环节，其效率与安全性对整个采矿过程至关重要。然而，露天采矿往往面临复杂多变的恶劣环境条件，如极端气候、光照不足、粉尘弥漫等，这些因素不仅显著降低了岩石破碎的效率，还大幅增加了发生安全事故的概率。在此背景下，本文提出一种基于改进YOLOv8n（You Only Look Once）的恶劣环境下岩石检测方法。该方法的核心之处在于：一是在YOLOv8n的主干网络的浅层引入差分边缘增强模块（Detail Enhancement，DE），通过增强检测目标的边缘信息，有效突出岩石在复杂背景中的轮廓特征；二是通过矩阵化颜色校正模块（Matrix-based Color Correction，MCC），构建全局颜色变换矩阵，对输入特征图进行全局颜色变换与局部细节优化，进一步提升特征表达的鲁棒性与准确性。实验结果表明，本文提出的方法能够显著提升恶劣环境下的岩石检测精度，为露天采矿中的岩石破碎作业提供了一种高效、可靠的检测技术。

Abstract

Rock fragmentation is a crucial process in the mining industry for extracting valuable minerals

and its efficiency and safety are of vital importance to the entire mining operation. However

open-pit mining environments are often confronted with complex and harsh conditions

such as extreme weather

insufficient lighting

and dust-laden atmospheres. These factors not only significantly reduce the efficiency of rock fragmentation but also greatly increase the probability of safety accidents. Against this backdrop

this paper proposes an improved YOLOv8n (You Only Look Once) algorithm for rock detection under harsh conditions. The core innovations of this algorithm include: (1) the introduction of a differential edge enhancement module in the shallow layers of the YOLOv8n backbone network to enhance the edge information of the detection targets and effectively highlight the contour features of rocks in complex backgrounds; (2) the construction of a global color transformation matrix using a matrix-based color correction module to perform global color transformation and local detail optimization on the input feature maps

thereby further improving the robustness and accuracy of feature representation. Experimental results demonstrate that the proposed method can significantly enhance the precision of rock detection under harsh conditions

providing an efficient and reliable detection technology for rock fragmentation operations in open-pit mining.

关键词

Keywords

references

刘送永 , 李洪盛 , 江红祥 , 等 . 矿山煤岩破碎方法研究进展及展望 [J ] . 煤炭学报 , 2023 , 48 ( 2 ): 1047 - 1069 .

罗会兰 , 陈鸿坤 . 基于深度学习的目标检测研究综述 [J ] . 电子学报 , 2020 , 48 ( 6 ): 1230 - 1239 .

江泽涛 , 施道权 , 雷晓春 , 等 . 一种基于Night-YOLOX的低照度目标检测方法 [J ] . 电子学报 , 2023 , 51 ( 10 ): 2821 - 2830 .

LAMPINEN S , NIU L C , HULTTINEN L , et al . Autonomous robotic rock breaking using a real-time 3D visual perception system [J ] . Journal of Field Robotics , 2021 , 38 ( 7 ): 980 - 1006 .

LIN M , CHEN Q , YAN S . Network in network [EB/OL ] . ( 2014-03-04 )[ 2024-05-23 ] . https://arxiv.org/abs/1312.4400 https://arxiv.org/abs/1312.4400 .

KINGMA D P , BA J . Adam: A method for stochastic optimization [EB/OL ] . ( 2017-01-30 )[ 2024-05-23 ] . https://arxiv.org/abs/1412.6980 https://arxiv.org/abs/1412.6980 .

LOSHCHILOV I , HUTTER F . SGDR: Stochastic gradient descent with warm restarts [EB/OL ] . ( 2016-08-01 )[ 2024-05-23 ] . http://arxiv.org/abs/1608.03983v3 http://arxiv.org/abs/1608.03983v3 .

WANG A , CHEN H , LIU L H , et al . YOLOv10: Real-time end-to-end object detection [EB/OL ] . ( 2024-05-23 )[ 2024-05-23 ] . https://arxiv.org/abs/2405.14458v2 https://arxiv.org/abs/2405.14458v2 .

ZHAO Y A , LYU W , XU S L , et al . Detrs beat yolos on real-time object detection [EB/OL ] . ( 2024-04-03 )[ 2024-05-23 ] . https://arxiv.org/abs/2304.08069v3 https://arxiv.org/abs/2304.08069v3 .

SUN P Z , ZHANG R F , JIANG Y , et al . Sparse R-CNN: An end-to-end framework for object detection [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 12 ): 15650 - 15664 .

BAO H , DONG L , PIAO S , et al . Beit: Bert pre-training of image transformers [EB/OL ] . ( 2022-09-03 )[ 2024-05-23 ] . https://arxiv.org/abs/2106.08254 https://arxiv.org/abs/2106.08254 .

CHEN S F , SUN P Z , SONG Y B , et al . DiffusionDet: Diffusion model for object detection [C ] // 2023 IEEE/CVF International Conference on Computer Vision (ICCV) . IEEE , 2023 : 19830 - 19843 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Neural Network Algorithm for Road Damage Detection Based on Feature Transfer

Enhanced Boundary-Free Discontinuous Rotation Detection Algorithm in YOLOv7

Related Author

XU Xi-yang

WANG Long-hui

YING Kun-yu

GUO Xiao-qiu

WAN Ji-peng

GUO Zhen-jiang

HE Xiao-hai

WU Xiao-hong

Related Institution

Jiangxi Tianchi Expressway Technology Development Co., Ltd.

Jiangxi Vocational and Technical College of Transportation

Nanchang University of Aeronautics and Astronautics

Tianjin City Construction University

College of Electronics and Information Engineering， Sichuan University

⁰