Journal of Sensor Science and Technology (센서학회지)

The Korean Sensors Society (한국센서학회)
권호 표지
DOI QR코드

DOI QR Code

A Design of Multilayer Perceptron for Camera Calibration

  • Do, Yong-Tae (School of Computer and Communication Engineering, Daegu University)
  • 도용태 (대구대학교 정보통신공학부)
  • Published : 2002.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper a new design of multi-layer perceptron(MLP) for camera calibration is proposed. Most existing techniques determine a transformation from 3D spatial points to their image points and camera parameters are tried to be estimated from the transformation. The technique proposed here, on the other hand, determines rays of sight uniquely from image points and parameters are estimated from the relationship using an MLP. By this approach projection and back-projection can be done more straightforwardly. Being based on a geometric model, a network design process becomes less ambiguous, which is a clear merit compared to other neural net based techniques. An MLP designed according to the technique proposed showed fast and stable learning in tests under various conditions.

Keywords

References

  1. Sensor Devices and Systems for Robotics, NATO ASI Series v.F52 The calibration problem for stereoscopic vision O. D. Faugueras;G. Toscani
  2. J. Korean Sensors Society v.7 no.6 Calibrating sterescopic 3D. position measurement systems using artificial neural nets Y. Do(et al.)
  3. IEEE Trans. Pattern Analysis and Machine Intelligence v.16 no.5 Implicit and explicit camera calibration: theory and experiments G-Q.Wei;S.D.Ma https://doi.org/10.1109/34.291450
  4. Advanced Robotics v.5 no.3 Robot vision modelling - camera modelling and camera calibration M. Ito https://doi.org/10.1163/156855391X00232
  5. IEEE J. Robotics & Automation v.RA-3 no.4 A versatile camera calibration technique for high accuracy 3d machine vision metrology using off-the-shelf TV cameras and lensens R. Y. Tsai
  6. Neural Networks v.3 no.5 Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks K. M. Hornik(et al.) https://doi.org/10.1016/0893-6080(90)90005-6
  7. Proc. Int. Joint Conf. Neural Networks Hybrid calibration of CCD cameras using artificial neural nets J. Wen;G. Schweitzer
  8. U.S. Patent, No.5617490 Camera system with neural network compensator for measuring 3-D position M. Kume;T.Kanade
  9. Int. J. Neural Systems v.8 no.3 Real-time neural network based camera localization and its extension to mobile robot control D-H.Choi;S-Y.Oh https://doi.org/10.1142/S012906579700029X
  10. Proc. TENCON Robust camera calibration using neural network J. Jun;C.Kim
  11. Proc. Int. Joint Conf. Neural Networks Application of neural networks for stereo-camera calibration Y. Do
  12. Proc. IEEE Int. Conf. Robotics and Automation Automatic training of a neural net for active stereo 3D reconstruction J. Neubert(et al.)
  13. Proc. Int. Conf. Image Processing A neural approach for single-and multi-image camera calibration M. Ahmed(et al.)
  14. Proc. IEEE Int. Conf. Robotics and Automation Decomposition of transformation matrices for robot vision S. Ganapathy
  15. IEEE Trans. Instrumentation and Measurement v.46 no.5 A three-step camera calibration method H. Bacakoglu;M.Kamel https://doi.org/10.1109/19.676732