3D shape measurement of translucent objects based on Fourier single-pixel imaging in projector-camera system

Area of Science:

• Optics and Photonics
• Computer Vision
• Metrology

Background:

• Structured light 3D shape measurement is widely used but struggles with translucent objects due to subsurface scattering.
• Global illumination effects, like scattering, introduce errors in conventional 3D shape recovery.
• Accurate 3D reconstruction of translucent objects remains a significant challenge in optical metrology.

Purpose of the Study:

• To develop a novel 3D shape measurement method for translucent objects.
• To address the limitations of structured light techniques when applied to scattering materials.
• To improve the accuracy and feasibility of 3D shape measurement for translucent objects.

Main Methods:

• Proposed a 3D shape measurement method based on Fourier single-pixel imaging (FSI).
• Utilized stereo matching to reconstruct 3D shapes from separated direct illumination light.
• Separated direct illumination light from subsurface scattering light using the FSI technique.

Main Results:

• Successfully separated direct illumination light from subsurface scattering light.
• Demonstrated the feasibility and accuracy of the proposed FSI-based method.
• Provided qualitative and quantitative results validating the method's performance.

Conclusions:

• The proposed Fourier single-pixel imaging method effectively measures the 3D shapes of translucent objects.
• The technique overcomes limitations of traditional structured light by separating scattering effects.
• This advancement offers a more accurate approach for 3D shape measurement of challenging translucent materials.