Transactions of the Korean Society of Mechanical Engineers A (대한기계학회논문집A)

The Korean Society of Mechanical Engineers (대한기계학회)
권호 표지
DOI QR코드

DOI QR Code

3D Object Recognition and Accurate Pose Calculation Using a Neural Network

인공신경망을 이용한 삼차원 물체의 인식과 정확한 자세계산

  • Published : 1999.11.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents a neural network approach, which was named PRONET, to 3D object recognition and pose calculation. 3D objects are represented using a set of centroidal profile patterns that describe the boundary of the 2D views taken from evenly distributed view points. PRONET consists of the training stage and the execution stage. In the training stage, a three-layer feed-forward neural network is trained with the centroidal profile patterns using an error back-propagation method. In the execution stage, by matching a centroidal profile pattern of the given image with the best fitting centroidal profile pattern using the neural network, the identity and approximate orientation of the real object, such as a workpiece in arbitrary pose, are obtained. In the matching procedure, line-to-line correspondence between image features and 3D CAD features are also obtained. An iterative model posing method then calculates the more exact pose of the object based on initial orientation and correspondence.

Keywords

References

  1. 대한기계학회논문집 v.19 no.12 로봇 아크용접 공정을 위한 용접접합부의 시각인식 및 추종 손영탁;김재선;조형석;고광일
  2. Neural Networks v.2 A Neural Network Approach to Character Recognition Rajavelu, A.;Musavi, M. T.;Shirvaikar, M. V.
  3. Pattern Recognition v.23 no.6 A Neural Network Approach to Robust Shape Classification Gupta, L.;Sayen, M. R.;Tammana, R.
  4. IEEE Transaction on Systems, Man, and Cybernetics v.21 no.6 Object Recognition by a Hopfield Neural Network Nasrabadi, N. M.;Li, W.
  5. Pattern Recognition v.25 no.1 Object Recognition using Invariant Object Boundary Representations and Neural Network Models Bebis, G. N.;Papadourakis, G. M.
  6. IEEE SOUTHEASTCON '93 An Approach to Three-Dimensional Object Recognition Using a Hybrid Hopfield Network Brooks, T. D.;Kim, J. H.
  7. Intenational Joint Conference on Neural Networks v.3 3D Object Recognition and Shape Estimation from Image Contours using B-Splines, Unwarping Techniques, and Neural Network Wang, J. Y.;Cohen, F. S.
  8. Nature v.343 no.18 A Network That Learns to Recognize Three-dimensional Objects Poggio, T.;Edelman, S.
  9. International Journal of Computer Vision v.5 no.2 Recognizing Solid Object by Alignment with an Image Huttenlocher, D. P.;Ullman, S.
  10. Computer Vision Ballard, D. H.;Brown, C. M.
  11. Artificial Intelligence v.31 Three-Dimensional Object Recognition from Single Two-Dimensional Images Lowe, D. G.
  12. Image and Vision Computing v.11 no.1 Pose Determination of Parameterized Object Models from a Monocular Image Goldberg, R. R.
  13. Digital Image Processing Gonzalez, R. C.;Woods, R. E.