Self-registration shape measurement based on fringe projection and structure from motion

Area of Science:

• Optics and Photonics
• Computer Vision
• Metrology

Background:

• Fringe-projection-based 3D reconstruction is a key technique for shape measurement.
• Patchwise point cloud registration remains a challenge without external aids.
• Existing methods often require rigidly connected camera-projector systems.

Purpose of the Study:

• To develop a flexible and robust self-registration method for 3D shape measurement.
• To overcome limitations of traditional structured-light devices.
• To enable independent movement of camera and projector for enhanced flexibility.

Main Methods:

• A novel image-capturing scheme and image-matching strategy are proposed.
• Structure from Motion (SfM) framework is utilized to calculate and optimize camera/projector positions.
• Sparse correspondence points act as virtual markers for global positioning.
• Dense 3D points are reconstructed using forward intersection.

Main Results:

• The proposed method achieves self-registration of 3D point clouds.
• It demonstrates comparable accuracy to state-of-the-art techniques.
• The system requires only one camera and one projector.
• No post-registration procedures or assistant markers are needed.

Conclusions:

• The developed method offers a flexible and robust solution for 3D shape measurement.
• It eliminates the need for rigid camera-projector connections and external markers.
• This approach simplifies the 3D reconstruction workflow and enhances applicability.