High dynamic range 3D measurements based on space-time speckle correlation and color camera

Area of Science:

• Optics and Photonics
• Computer Vision
• Metrology

Background:

• Structured light (SL) 3D measurement challenges in high dynamic range (HDR) scenes with simultaneous high and low reflectivity.
• Existing methods struggle with scene complexity and require extensive data acquisition.

Purpose of the Study:

• To develop an efficient and accurate 3D measurement method for HDR scenes using SL.
• To enhance the dynamic range adaptability and reconstruction quality of SL systems.

Main Methods:

• Joint hardware and algorithm optimization for SL 3D measurement.
• Temporal projection of two sets of complementary speckle patterns for height encoding.
• Stereo matching strategy using space-time binary feature (ST-BIF) descriptor and zero-mean normalized cross-correlation (ST-ZNCC).
• Fusion of R, G, and B channel disparities to enhance measurable dynamic range.

Main Results:

• Achieved a root mean square error (RMSE) of 0.2516mm, significantly outperforming common ZNCC (1.0668mm).
• Attained an average coverage rate of 94.87%, compared to 93.35% with common ZNCC.
• Demonstrated successful 3D reconstruction of HDR scenes, including specular reflection regions, with only four frames.

Conclusions:

• The proposed method effectively addresses SL limitations in HDR scene 3D measurement.
• The joint hardware-algorithm design and advanced stereo matching significantly improve accuracy and coverage.
• This technique offers a robust solution for complex 3D metrology applications.