Optical 3-D surface reconstruction with color binary speckle pattern encoding.

Area of Science:

• Optics and Photonics
• Metrology
• Computer Vision

Background:

• Traditional 3-D surface measurement often relies on complex setups or multiple captures.
• Binary speckle pattern (BSP) methods offer advantages but can be limited in resolution and speed.
• Dynamic object measurement presents unique challenges due to motion blur and temporal synchronization.

Purpose of the Study:

• To develop a novel 3-D surface profile measurement scheme using a single-shot color binary speckle pattern (CBSP).
• To enhance measurement accuracy, spatial resolution, and efficiency for both static and dynamic objects.
• To establish a robust temporal-spatial correlation matching algorithm for 3-D reconstruction from CBSP.

Main Methods:

• A single-shot color binary speckle pattern (CBSP) was designed, coupling R/G/B channels with three black and white binary speckle patterns (BWBSPs).
• Mathematical derivations established the relationship between CBSP design and system configuration parameters.
• An extended temporal-spatial correlation framework was developed for stereo image sequence matching and 3-D reconstruction.

Main Results:

• Comparative experiments demonstrated superior spatial resolution, accuracy, and efficient reconstructed points compared to conventional BWBSP methods.
• The CBSP method successfully measured the 3-D profile of a moving A4 paper, validating its capability for dynamic objects.
• Standard workpieces (dumbbell, optical flat) were used to assess and confirm measurement accuracy.

Conclusions:

• The proposed CBSP-based 3-D surface profile measurement scheme offers significant advantages over existing methods.
• The developed temporal-spatial correlation algorithm effectively handles stereo image sequences for accurate 3-D reconstruction.
• This novel approach provides a promising solution for high-resolution, efficient 3-D measurement of dynamic and static objects.