Rock Identification Method based on YOLOv8
DOI:
https://doi.org/10.6919/ICJE.202408_10(8).0016Keywords:
YOLOv8; Rock Recognition; Deep Learning; Object Detection; Geological Exploration.Abstract
The traditional rock identification methods mainly rely on the field survey and laboratory analysis of geologists. These methods are not only time-consuming and labor-intensive, but also limited by the experience and subjective judgment of geologists. With the rapid development of computer vision and deep learning technology, especially the continuous progress of target detection algorithm, automatic rock recognition technology based on image processing has gradually become a research hotspot. In this paper, a rock recognition system based on YOLOv8 is proposed. The system utilizes the efficiency and accuracy of YOLOv8 algorithm to realize the fast and accurate recognition of various rock types. By constructing a rock image dataset, training a deep learning model, and testing it in a real scene, the system achieves satisfactory recognition results. The research in this paper is of great significance for improving the efficiency and accuracy of rock identification and promoting geological exploration and resource development.
Downloads
References
Xu Shenglin, Shao Zhaogang, Li Bing, et al. Electrical structures in western qaidam basin lithosphere studies [J/OL]. Geological bulletin, 1-16 [2024-07-11]. http://111.229.208.102:8085/kcms/detail/ 11.46 48.P.20240709.1149.002.html.
Yuan Ye, Zhou Shuwei. Experimental method and feasibility study of Type I fracture of rocks subjected to cyclic thermal action [J]. Municipal Technology, 2019,42(07):1-6+21.DOI:10.19922/j.1009-7767. 2024.07.001.
Wang Shiquan, Peng Jiangtao, Ye Ping, et al. Geochemical characteristics and prospecting potential analysis of the Beijiashan molybdenum gold deposit, Henan Province [J/OL]. Mineral exploration, 1-10 [2024-07-11]. http://111.229.208.102:8085/kcms/detail/11.5875.TD.20240701.0947.010.html.
Chen Hongcan, Xie Chaoming, Zhangjiajun, et al. Zircon U-Pb dating, petrogenesis and implications for Proto-Tethys Ocean subduction of Late Cambian-Early Ordovician rhyolite in Zhapeng area, Lhasa Block [J/OL]. Geological bulletin, 1-22 [2024-07-11]. http://111.229.208.102:8085/kcms/detail/11.4648. p.20240705.1123.004.html.
Lu Xiaoyu, Xie Ya, Dong Chunliang, et al. Single fracture rock creep mechanics property and damage evolution law [J/OL]. Journal of applied mechanics, 1-14 [2024-07-11]. http://111.229.208.102: 8085/kcms/detail/61.1112.O3.20240627.1553.004.h tml.
Liu XC, Gao H Y, Qin Dewen, et al. Application of geostress-rock mechanics analysis in horizontal well fracturing of low permeability tight sandstone gas reservoir in East China Sea [J]. World oil industry, 2024, 31 (3) : 78-89. The DOI: 10.20114 / j.i SSN. 1006-0030.20240429001.
Zhao Chunchen, Liu Huchuang, Chang Bin, et al. Petrographic characteristics and oil and gas exploration indications of the Shanbei Slope reservoir in Ordos Basin: A case study of the Beishang6 oil Formation in Youfangzhuang [J]. Geology and resources, 2024 (03) : 331-338 + 309. DOI: 10.13686 / j.carol carroll nki dzyzy. 2024.03.008.
Tong Rong-Chao.(2023). Master of Lithology identification by Hyperspectral method combined with Machine Learning (Dissertation, Xi 'an Shiyou University). Master of https://link.cnki.net/doi/10.27400/ d.cnki.gxasc.2023.000839doi:10.27400/d.cnki.gxasc.2023.000839.
Wang H Y. (2020). Research and Application of intelligent identification of rock species based on image object detection technology, M.S. (Dissertation, Northeastern University). Master of https://link. cnki.net/doi/10.27007/d.cnki.gdbeu.2020.001449doi:10.27007/d.cnki.gdbeu.2020.001449.
Lei Mingfeng, Zhang Yunbo, Wang Weidong, et al. Rock lithology Mask R - CNN intelligent identification method and the application study [J]. Journal of railway science and engineering, 2022, 12 (11) : 3372-3382. The DOI: 10.19713 / j.carol carroll nki/u.t 20211473. 43-1423.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
