Research of Camera Calibration Based on BP Neural Network
Camera calibration is key technologies of three-dimensional image reconstruction, the traditional binocular camera calibration algorithm needs to create complex mathematical models, while the neural network can effectively handle non-linear mapping problem. In this paper, a method of camera calibration based on BP neural network is proposed, which well describe the non-linear relationship between the characteristics in the three-dimensional point coordinates and the corresponding pixel camera among binocular vision. The experimental results show that neural network-based calibration method for binocular vision can get a higher calibration accuracy.
KeywordsThree-dimensional reconstruction stereoscopic vision BP neural network camera calibration
Unable to display preview. Download preview PDF.
- 1.Xu, G., Zhang, Z.: Epipolar geometry in stereo, motion and object recognition. A unified approach. Kluwer Acamdemic Publisher (1996)Google Scholar
- 2.Hartly, R., Zisserman, A.: Multiple View Geometry in Computer Vision. Cambridge University Press (2003)Google Scholar
- 3.Tsai, R.: An efficient and accurate camera calibration technique for 3D machine vision. In: Proceedings of Computer Vision and Pattern Recognition, pp. 364–374 (1986)Google Scholar
- 4.Zhang, Z.: Flexible camera calibration by viewving a plane form unknown orientations. In: Proceedings of the 7th International Conference on Computer Vision, Greece, pp. 666–673 (1999)Google Scholar
- 5.Meng, X.Q., Li, H., Hu, Z.Y.: A new camera calibration technique based on circular points. In: Proceedings of British Machine Vision Conference, U.K., pp. 496–505 (September 2000)Google Scholar
- 8.Yang, J.P., Li, Q., Liu, Z., Yuan, X.L.: Research of improved BP algorithm based on self-adaptive learning rate. Computer Engineering and Applications 45(11), 56–58 (2009)Google Scholar
- 9.Martin, T.H., Howard, B.D., Mark, H.B.: Neural network design, pp. 197–222. Mechanical Industry Press (2002)Google Scholar