基于双目立体视觉的电解阴极铜板三维重建研究
Research on 3D Reconstruction of Electrolytic Cathode Copper Plate Based on Binocular Stereo Vision
DOI: 10.12677/MOS.2023.125399, PDF,   
作者: 王诗杰, 袁嫣红*:浙江理工大学机械工程学院,浙江 杭州;鄢和平:浙江勇峰智能科技有限公司,浙江 绍兴
关键词: 双目立体视觉精度分析立体匹配三维重建Binocular Stereo Vision Precision Analysis Stereo Matching 3D Reconstruction
摘要: 针对电解生产阴极铜过程中对其表面质量监测的需求,提出基于双目立体视觉技术进行阴极铜表面具有深度信息瑕疵三维重建方法。首先通过分析大视场下平行双目检测精度模型,建立具有宽基线的双目立体结构,保证定位理论精度。然后为了进行基于SGBM的三维重建,利用稀疏特征点匹配缩小视差搜索范围,提升宽基线稠密立体匹配的效率及精度。使用设计系统进行测量实验,结果表明:该方法实现了对具有瑕疵的电解阴极铜板误差1 mm内的三维重建。这对于电解生产过程中的质量监测具有重要意义,可以帮助及时发现和定位表面瑕疵,从而提高产品质量并减少生产成本。
Abstract: Aiming at the requirement of monitoring the surface quality of cathode copper in the process of electrolytic production, a three-dimensional reconstruction method based on binocular stereo vi-sion technology for defects with depth information on the surface of cathode copper is proposed. First, by analyzing the parallel binocular detection accuracy model under a large field of view, a binocular stereo structure with a wide baseline is established to ensure the theoretical accuracy of positioning. Then, in order to perform 3D reconstruction based on SGBM, sparse feature point matching is used to narrow the disparity search range and improve the efficiency and accuracy of wide baseline dense stereo matching. Using the design system to carry out measurement experi-ments, the results show that: the method realizes the three-dimensional reconstruction of the elec-trolytic cathode copper plate with defects within 1 mm error. This is of great significance for quality monitoring in the electrolytic production process, and can help to detect and locate surface defects in time, thereby improving product quality and reducing production costs.
文章引用:王诗杰, 鄢和平, 袁嫣红. 基于双目立体视觉的电解阴极铜板三维重建研究[J]. 建模与仿真, 2023, 12(5): 4379-4392. https://doi.org/10.12677/MOS.2023.125399

参考文献

[1] 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. GB/T467-2010, 阴极铜[S]. 北京: 中国标准出版社, 2010.
[2] 丁少闻, 张小虎, 于起峰, 等. 非接触式三维重建测量方法综述[J]. 激光与光电子学进展, 2017, 54(7): 27-41.
[3] 袁红超. 基于机器视觉的电解阴极铜分拣系统的研发[D]: [硕士学位论文]. 昆明: 昆明理工大学, 2018.
[4] Gu, F., Song, Z. and Zhao, Z. (2020) Single-Shot Structured Light Sensor for 3D Dense and Dynamic Reconstruc-tion. Sensors, 20, 1094. [Google Scholar] [CrossRef] [PubMed]
[5] Zhang, Y., Liu, W., Lu, Y.K., Cheng, X.K., Luo, W.Q., Di, H.T. and Wang, F.J. (2020) Accurate Profile Measurement Method for Industrial Stereo-Vision Systems. Sensor Review, 40, 445-453. [Google Scholar] [CrossRef
[6] 马浩. 典型立体匹配算法精化方法研究[D]: [博士学位论文]. 武汉: 武汉大学, 2018.
[7] 乌彩妮, 王国珲, 杜哲琪. 一种基于光束平差法的双目鱼眼系统标定方法[J]. 西安工业大学学报, 2023, 43(1): 32-40.
[8] 肖志涛, 张文寅, 耿磊, 张芳, 吴骏. 双目视觉系统测量精度分析[J]. 光电工程, 2014, 41(2): 6-11.
[9] Hirschmüller, H. (2008) Stereo Processing by Semiglobal Matching and Mutual Information. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30, 328-341. [Google Scholar] [CrossRef
[10] Rublee, E., Ra-baud, V., Konolige, K. and Bradski, G.R. (2011) ORB: An Efficient Alternative to SIFT or SURF. 2011 International Confer-ence on Computer Vision, Barcelona, 6-13 November 2011, 2564-2571. [Google Scholar] [CrossRef
[11] Birchfield, S. and Tomasi, C. (1998) Depth Discontinuities by Pix-el-to-Pixel Stereo. Sixth International Journal of Computer Vision, Bombay, 7-7 January 1998, 1073-1080. [Google Scholar] [CrossRef
[12] Zhang, Z. (2000) A Flexible New Technique for Camera Calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22, 1330-1334. [Google Scholar] [CrossRef
[13] Li, J., Wang, P., Xiong, P., Cai, T., Yan, Z., Yang, L., Liu, J., Fan, H. and Liu, S. (2022) Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlation. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, 18-24 June 2022, 16242-16251. [Google Scholar] [CrossRef