Guided, Fusion-Based, Large Depth-of-field 3D Imaging Using a Focal Stack

Area of Science:

• Optics and Photonics
• Computer Vision and Image Processing

Background:

• Structured light 3D imaging systems face limitations in working volume due to aperture requirements.
• Achieving simultaneous focus for both projector and camera restricts depth-of-field in active 3D imaging.

Purpose of the Study:

• To propose a novel method for extending the working range of structured light 3D imaging systems.
• To enable high-quality 3D shape measurement in scenes with large depth variations.

Main Methods:

• Utilized a focal stack approach combined with the gray code method for local 3D shape measurement.
• Generated a global coarse depth map by extracting texture maps from multiple focal positions.
• Fused local measurements guided by the coarse depth map for comprehensive scene reconstruction.

Main Results:

• Developed a prototype system demonstrating high-quality 3D measurements.
• Achieved a measurement depth range of 400 mm.
• Reported a low measurement error of 0.08%.

Conclusions:

• The proposed focal stack method effectively extends the working range of structured light 3D imaging.
• This technique enables accurate 3D reconstruction in challenging scenes with significant depth variations.
• The system offers a practical solution for industrial measurements and cultural heritage preservation.