A new calibration method for line-structured light vision sensors based on concentric circle feature

Mingwei Shao, Junyu Dong, Amanuel Hirpa MadessaJournal of the European Optical Society Rapid Publications 2019 · DOI

Abstract

Determining the relative ubiety between the camera and the laser projector in a line-structured light vision sensor is a classical yet important task. Typical calibration methods often confront problems, such as difficulty of producing the target precisely and introduction of perspective projection errors. In this work, a new calibration method based on a concentric circle feature is introduced. The proposed method is based on geometrical properties and can reduce the perspective projection error. In our method, the vanishing line of the light plane is firstly deduced from the imaged concentric circles. Then the normal vector of the light plane is determined. Consequently, the complete expression can be confirmed from the principle of the intersecting planes. The proposed method is simple and robustness as the basic theory is geometrical properties. Accuracy evaluation experiment shows that the accuracy of the calibration method can reach 0.07 mm within the view field of about 200 × 200 mm. This accuracy is comparable to the commonly used calibration method with a checkerboard planar target, whereas our target is simple to produce.

Introduction

Line-structured light vision sensors (LSLVS) are essential tools in industrial 3D measurement. These systems typically consist of a camera and a laser projector that projects a line onto an object's surface. To derive accurate 3D measurements, the geometric relationship between the camera and the laser "light plane" must be precisely determined through a process known as calibration.

Geometric model of a line-structured light vision sensor.

Traditional calibration methods often face significant trade-offs. High-accuracy 3D targets are expensive and difficult to manufacture. 2D checkerboards, while popular, are prone to "perspective projection errors," where the imaged center of a feature does not align with its physical center due to the camera's viewing angle. 1D targets offer convenience but often lack the necessary precision for high-end industrial applications. This paper introduces a method that combines the manufacturing simplicity of 1D targets with the high precision of 2D/3D methods by utilizing concentric circle features.

Methodology

The proposed method leverages the projective geometry of concentric circles to minimize perspective-related errors. The core of the approach involves extracting the vanishing line of the target plane directly from the imaged ellipses of the concentric circles.

Concentric circle target geometry: (a) top-down view and (b) perspective view on the target plane.
  1. Target Design: The calibration target is a planar board featuring two or more concentric circles.
  2. Vanishing Line Extraction: When viewed at an angle, the circles appear as ellipses in the image. By representing these ellipses as matrices and performing an eigen-decomposition, the vanishing line of the target plane can be mathematically deduced. This geometric approach is inherently more robust against perspective deviations than point-based matching.
  3. Normal Vector Determination: The extracted vanishing line allows for the calculation of the target plane's normal vector within the camera's coordinate system.
  4. Light Plane Calculation: As a laser stripe intersects the target at various positions, multiple vanishing points are collected. These points define the vanishing line of the laser light plane itself.
  5. Final Calibration: Using the principle of intersecting planes, the complete mathematical equation for the laser light plane is established.

Experimental Results

The method was validated through both computer simulations and real-world experiments. Simulations demonstrated that increasing the number of concentric circles (from two to four) significantly improves robustness against image noise by providing more constraints for vanishing line estimation.

In physical experiments using an AVT F-504B camera and a standard line laser, the system achieved a Root Mean Square (RMS) error of 0.072 mm within a 200 × 200 mm field of view. This performance is highly competitive, matching the accuracy of 2D checkerboard methods (0.070 mm) and 1D methods (0.065 mm), while significantly outperforming traditional 3D target methods (0.151 mm) in the same setup.

Conclusion

The concentric circle-based calibration method offers a robust and accurate alternative for industrial LSLVS applications. Its primary advantages are the simplicity of the target manufacturing and its resilience to perspective projection errors. While the method requires moving the target to several orientations to gather sufficient data, its precision and ease of use make it suitable for on-site industrial calibration. Future work may focus on automating target manipulation to further streamline the process.

Summary updated

Abstract

Determining the relative ubiety between the camera and the laser projector in a line-structured light vision sensor is a classical yet important task. Typical calibration methods often confront problems, such as difficulty of producing the target precisely and introduction of perspective projection errors. In this work, a new calibration method based on a concentric circle feature is introduced. The proposed method is based on geometrical properties and can reduce the perspective projection error. In our method, the vanishing line of the light plane is firstly deduced from the imaged concentric circles. Then the normal vector of the light plane is determined. Consequently, the complete expression can be confirmed from the principle of the intersecting planes. The proposed method is simple and robustness as the basic theory is geometrical properties. Accuracy evaluation experiment shows that the accuracy of the calibration method can reach 0.07 mm within the view field of about 200 × 200 mm. This accuracy is comparable to the commonly used calibration method with a checkerboard planar target, whereas our target is simple to produce.
BibTeX
@article{shao2019new,
  author = {Mingwei Shao and Junyu Dong and Amanuel Hirpa Madessa},
  title = {{A new calibration method for line-structured light vision sensors based on concentric circle feature}},
  journal = {Journal of the European Optical Society Rapid Publications},
  year = {2019},
  doi = {10.1186/s41476-019-0097-0},
}
Figure