Valid point detection in fringe projection profilometry.

Area of Science:

• Optics and Photonics
• Computer Vision and Image Processing
• Metrology

Background:

• Fringe projection profilometry is a widely used 3D imaging technique.
• Existing methods for valid point detection face challenges in segmentation, error correction, and noise removal.
• Current techniques often require inconvenient, case-dependent thresholds.

Purpose of the Study:

• To develop a robust and user-friendly framework for valid point detection in fringe projection profilometry.
• To overcome limitations of existing methods by automating threshold setting and improving accuracy.
• To enhance the reliability of 3D data acquisition using fringe projection techniques.

Main Methods:

• Utilized k-means clustering for automated background segmentation.
• Implemented theoretical analysis for unwrapping error correction.
• Developed temporal and spatial noisy point detection with automatic threshold determination.

Main Results:

• The proposed framework successfully performs automatic background segmentation.
• Unwrapping errors are effectively corrected based on theoretical analysis.
• Noisy points are accurately detected in both temporal and spatial domains with automated thresholds.

Conclusions:

• The novel framework offers a significant improvement for valid point detection in fringe projection profilometry.
• Automatic threshold setting enhances processing convenience and applicability.
• The validated framework contributes to more reliable and efficient 3D information acquisition.