General object recognition is difficult. We had proposed a novel solution for object recognition in an unconstrained environment using a tag (Matas[2]). This simplifies the problem by placing tags of special pattern on the objects that allows us to determine the pose easily. A robust calibration chart detector was developed for the first stage of the solution (Soh[1]). This paper investigates the next part of the solution, i.e. the model acquisition and matching using the Chamfer matching algorithm. The algorithm is reasonably simple to implement and very efficient in terms of computation. We experiment with this technique extensively to prove the reliability of the approach. Using this approach, the objective of Tagged Object Recognition (TOR) can be realised and we should be able to perform object recognition wherever a tag is located. The technology will facilitate applications in landmark and object recognition, mobile robot navigation and scene modelling.
[1]
Roger Tsai,et al.
Synopsis of recent progress on camera calibration for 3D machine vision
,
1989
.
[2]
Gunilla Borgefors,et al.
Hierarchical Chamfer Matching: A Parametric Edge Matching Algorithm
,
1988,
IEEE Trans. Pattern Anal. Mach. Intell..
[3]
Roger Y. Tsai,et al.
A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses
,
1987,
IEEE J. Robotics Autom..
[4]
Josef Kittler,et al.
Robust recognition of calibration charts
,
1997
.
[5]
Jiri Matas,et al.
Object recognition using a tag
,
1997,
Proceedings of International Conference on Image Processing.