首页 > 论文 > 激光与光电子学进展 > 56卷 > 11期(pp:111505--1)

复杂环境下基于深度神经网络的摄像机标定

Camera Calibration Based on Deep Neural Network in Complex Environments

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

提出一种基于深度神经网络的摄像机标定方法,实现了复杂环境下平面区域内的灵活、高精度标定。无需进行数据特征提取或分类,仅通过优化网络结构、超参数与训练算法,深度神经网络便能得到快速有效的训练。实验结果表明,相较于张正友标定法与浅层神经网络标定法,该方法在大范围、多拍摄角度和高畸变条件下均能达到更高的标定精度,镜头存在高畸变时,633 mm×763 mm标定范围内的平均标定误差仅为0.1471 mm。

Abstract

This study proposes a new deep neural network based camera calibration method that achieves flexible, high-precision calibration in complex environments, without having to classify or extract features from input data. By optimizing the network structure, hyperparameters, and training algorithms, the deep neural network can be quickly and effectively trained. The experimental results confirm that, compared with Zhang's calibration method and the shallow neural network, the proposed method can achieve high calibration accuracy under a wide range of imaging conditions involving multiple shooting angles or high distortion. For the images produced using a highly distorted lens, the proposed method achieves an average calibration error of only 0.1471 mm over the calibration range of 633 mm×763 mm.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/LOP56.111505

所属栏目:机器视觉

基金项目:国家重点研发计划、国家自然科学基金、国家自然科学基金青年基金;

收稿日期:2018-12-20

修改稿日期:2019-01-07

网络出版日期:2019-06-13

作者单位    点击查看

向鹏:东南大学能源与环境学院, 江苏 南京 210096
周宾:东南大学能源与环境学院, 江苏 南京 210096
祝仰坤:东南大学能源与环境学院, 江苏 南京 210096
贺文凯:东南大学能源与环境学院, 江苏 南京 210096
岳晓庚:东南大学能源与环境学院, 江苏 南京 210096
陶依贝:东南大学能源与环境学院, 江苏 南京 210096

联系人作者:周宾(seuxp@foxmail.com)

备注:国家重点研发计划、国家自然科学基金、国家自然科学基金青年基金;

【1】Zhang Z Y. Camera calibration. ∥Ikeuchi K, Computer Vision: A Reference Guide. Boston: Springer. 76-77(2014).

【2】Tsai R. A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses. IEEE Journal on Robotics and Automation. 3(4), 323-344(1987).

【3】Zhang Z. A flexible new technique for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence. 22(11), 1330-1334(2000).

【4】Zou P P, Zhang Z L, Wang P et al. Binocular camera calibration based on collinear vector and plane homography. Acta Optica Sinica. 37(11), (2017).
邹朋朋, 张滋黎, 王平 等. 基于共线向量与平面单应性的双目相机标定方法. 光学学报. 37(11), (2017).

【5】Wu Q H, Chen H, Zhu S S et al. Fast calibration method of large dimension measurement system based on multi-camera array. Acta Optica Sinica. 38(12), (2018).
吴庆华, 陈慧, 朱思斯 等. 一种多相机阵列大尺寸测量系统快速标定方法. 光学学报. 38(12), (2018).

【6】Zhu F, Yu F S, Wu Y M et al. Analysis of attitude calibration precision of P4P camera. Acta Optica Sinica. 38(11), (2018).
朱帆, 于芳苏, 吴易明 等. P4P法相机姿态标定精度分析. 光学学报. 38(11), (2018).

【7】Ramalingam S and Sturm P. A unifying model for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence. 39(7), 1309-1319(2017).

【8】Hornik K, Stinchcombe M and White H. Multilayer feedforward networks are universal approximators. Neural Networks. 2(5), 359-366(1989).

【9】Wen J and Schweitzer G. Hybrid calibration of CCD cameras using artificial neural nets. [C]∥1991 IEEE International Joint Conference on Neural Networks, November 18-21, 1991, Singapore. New York: IEEE. 337-342(1991).

【10】Jun J and Kim C. Robust camera calibration using neural network. Cat. No.99CH37030), September 15-17, 1999, Cheju Island, Korea. New York: IEEE. 694-697(1999).

【11】Memon Q and Khan S. Camera calibration and three-dimensional world reconstruction of stereo-vision using neural networks. International Journal of Systems Science. 32(9), 1155-1159(2001).

【12】Lü C H, Zhang Z Y and An P. Camera calibration for stereo vision based on neural network. Chinese Journal of Mechanical Engineering. 39(9), 93-96(2003).
吕朝辉, 张兆杨, 安平. 基于神经网络的立体视觉摄像机标定. 机械工程学报. 39(9), 93-96(2003).

【13】Tian Z, Zhang Q, Xiong J L et al. Large-scale camera calibration with neural network. Acta Optica Sinica. 31(4), (2011).
田震, 张玘, 熊九龙 等. 基于神经网络的大范围空间标定技术. 光学学报. 31(4), (2011).

【14】Liu X J, Li X J, Wang W Y et al. Large-scale binocular camera calibration combining neural network with virtual target. Optical Technique. 43(4), 314-318(2017).
刘小娟, 李学军, 王文韫 等. 基于神经网络的虚拟靶标大视场双目相机标定技术. 光学技术. 43(4), 314-318(2017).

【15】LeCun Y, Bengio Y and Hinton G. Deep learning. Nature. 521(7553), 436-444(2015).

【16】Hinton G, Deng L, Yu D et al. Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Processing Magazine. 29(6), 82-97(2012).

【17】Hinton G E, Osindero S and Teh Y W. A fast learning algorithm for deep belief nets. Neural Computation. 18(7), 1527-1554(2006).

【18】Glorot X, Bordes A and Bengio Y. Deep sparse rectifier neural networks. [C]∥ Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, April 11-13, 2011, Fort Lauderdale, FL, USA. Cambridge: PMLR. 15, 315-323(2011).

【19】Bergstra J and Bengio Y. Random search for hyper-parameter optimization. Journal of Machine Learning Research. 13(1), 281-305(2012).

引用该论文

Peng Xiang, Bin Zhou, Yangkun Zhu, Wenkai He, Xiaogeng Yue, Yibei Tao. Camera Calibration Based on Deep Neural Network in Complex Environments[J]. Laser & Optoelectronics Progress, 2019, 56(11): 111505

向鹏, 周宾, 祝仰坤, 贺文凯, 岳晓庚, 陶依贝. 复杂环境下基于深度神经网络的摄像机标定[J]. 激光与光电子学进展, 2019, 56(11): 111505

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF