Accurate stereo vision system calibration with chromatic concentric fringe patterns

Area of Science:

• Computer Vision
• 3D Imaging Systems
• Metrology

Background:

• Structured light 3D imaging systems rely on camera calibration for accurate intrinsic and extrinsic parameter determination.
• Traditional intensity-based calibration methods using patterns like chessboards are sensitive to image noise, blur, and contrast variations.
• Robust and accurate camera calibration is crucial for high-fidelity 3D reconstruction.

Purpose of the Study:

• To develop an active camera calibration method for structured light 3D imaging systems.
• To improve the accuracy and robustness of feature point detection in challenging imaging conditions.
• To enhance the overall performance of 3D reconstruction by minimizing calibration errors.

Main Methods:

• Proposed an active calibration technique utilizing chromatic concentric fringe patterns (CCFP).
• Employed a phase analysis algorithm to acquire circular phase information.
• Applied an ellipse fitting algorithm to extract subpixel center coordinates from phase contours for precise calibration.

Main Results:

• The proposed method demonstrates superior robustness to Gaussian blur and noise compared to traditional approaches.
• Simulations and experiments confirmed higher calibration accuracy.
• A reduction in reprojection error by at least 10% was observed under out-of-focus conditions.
• 3D reconstruction accuracy improved by nearly 10%.

Conclusions:

• The active calibration method using CCFP offers a more reliable approach for structured light 3D imaging.
• The technique significantly improves calibration accuracy and robustness, especially in adverse imaging conditions.
• This advancement contributes to more precise and reliable 3D reconstruction in various applications.