Color-encoded single-shot computer-generated Moiré profilometry

Hechen Zhang, Yiping Cao, Chengmeng Li, Lu Wang, Hongmei Li, Cai Xu, Yingying WanScientific 2021 · DOI

Abstract

A color-encoded single-shot computer-generated Moiré profilometry (CSCGMP) is proposed. Two sinusoidal gratings with a π phase difference are encoded in red and blue channels respectively to combine a composite color grating. While this composite color grating is projected onto the measured object, the corresponding color deformed pattern can be captured. So two deformed patterns with a π phase difference are separated from its red and blue components respectively. After normalization and subtraction, the AC component of both separated deformed patterns can be extracted. If this AC component respectively multiplied by the two AC components of fringe patterns of reference plane with a π/2 phase difference prepared and saved on the computer in advance, two computer-generated Moiré fringes just respectively standing for sine and cosine of phase which is modulated by the height of the object relative to the reference plane are figured out. So the 3D shape of the measured object can be reconstructed with normal computer-generated Moiré profilometry. Both simulation and experimental results show the feasibility and validity of the proposed method. It has potential in real-time 3D measurement due to its single-shot feature.

Introduction: The Speed-Accuracy Trade-off

In 3D optical metrology, researchers often face a classic trade-off between speed and accuracy. Phase Measuring Profilometry (PMP) offers high precision but requires multiple shots, making it unsuitable for dynamic scenes. Conversely, single-shot methods like Fourier-transform Profilometry (FTP) are fast but rely on frequency-domain filtering. This filtering often fails when an object's complex geometry causes the signal (AC component) and background light (DC component) to overlap in the spectrum, leading to significant measurement errors.

Computer-Generated Moiré Profilometry (CGMP) emerged as a solution, yet previous iterations still struggled with spectral overlap or required multiple images to maintain high precision. The proposed Color-encoded Single-shot Computer-generated Moiré Profilometry (CSCGMP) addresses these limitations by enabling high-precision 3D sensing in a single frame.

Methodology: Color-Encoded CSCGMP

The core innovation of CSCGMP lies in its spatial domain processing and hardware-aware color encoding. The workflow is divided into two primary phases: a preparing procedure and a measuring procedure.

Overview of the CSCGMP measuring and preparing procedures, illustrating the separation of color channels and the generation of virtual Moiré fringes.

1. Color Encoding and Separation

A composite color grating is designed where two sinusoidal patterns with a $\pi$ phase difference are encoded into the Red and Blue channels. These channels are selected based on the spectral transmittance curves of standard CCD sensors to minimize "cross-talk." When projected onto an object, the captured color image is separated back into its Red ($I_r$) and Blue ($I_b$) components.

2. Spatial Purification

Rather than using Fourier transforms, CSCGMP cancels out background light (the DC component) in the spatial domain. By normalizing and subtracting the separated channels, the system extracts a "purified" AC signal. This method is robust even when the frequency spectra of the signal and background overlap, preserving high-frequency details that traditional filtering might blur.

3. Virtual Moiré Generation and Reconstruction

The purified AC component is multiplied by two pre-stored reference patterns (sine and cosine) representing the reference plane. This generates "virtual" Moiré fringes ($I_{Moiré0}, I_{Moiré90}$). Standard phase-unwrapping and height-to-phase mapping are then applied to reconstruct the final 3D shape.

Experimental Validation and Results

The method was validated through both numerical simulations and physical experiments. In simulations using a mathematical "peaks" function, the system achieved a maximum error of 0.0652 mm and a Root Mean Square (RMS) error of 0.0892 mm.

Simulation results (a-g) demonstrating the 3D reconstruction process, from the initial height map and color-encoded fringe to the final reconstructed object and error distribution.

Physical testing involved static models (a star, a human face, and a textured turtle) and dynamic objects. The system successfully reconstructed a moving human hand at 42 frames per second (fps), with a processing time of approximately 0.0238 seconds per frame. Comparative benchmarks against an 8-step PMP "gold standard" confirmed that CSCGMP provides superior accuracy over traditional FTP and standard CGMP methods.

Discussion and Future Work

CSCGMP represents a significant step forward for real-time 3D sensing. By moving background removal from the frequency domain to the spatial domain via color channel subtraction, it maintains high accuracy without sacrificing speed.

However, a notable limitation remains: the "Color Object" challenge. Since the method relies on Red and Blue channels to carry phase information, objects with strong intrinsic colors (e.g., a bright red surface) can interfere with the projected patterns. Future research will focus on developing decoupling techniques to separate an object's intrinsic color from the projected fringe information, expanding the method's applicability to a wider range of real-world materials.

Summary updated

Abstract

A color-encoded single-shot computer-generated Moiré profilometry (CSCGMP) is proposed. Two sinusoidal gratings with a π phase difference are encoded in red and blue channels respectively to combine a composite color grating. While this composite color grating is projected onto the measured object, the corresponding color deformed pattern can be captured. So two deformed patterns with a π phase difference are separated from its red and blue components respectively. After normalization and subtraction, the AC component of both separated deformed patterns can be extracted. If this AC component respectively multiplied by the two AC components of fringe patterns of reference plane with a π/2 phase difference prepared and saved on the computer in advance, two computer-generated Moiré fringes just respectively standing for sine and cosine of phase which is modulated by the height of the object relative to the reference plane are figured out. So the 3D shape of the measured object can be reconstructed with normal computer-generated Moiré profilometry. Both simulation and experimental results show the feasibility and validity of the proposed method. It has potential in real-time 3D measurement due to its single-shot feature.
BibTeX
@inproceedings{zhang2021color,
  author = {Hechen Zhang and Yiping Cao and Chengmeng Li and Lu Wang and Hongmei Li and Cai Xu and Yingying Wan},
  title = {{Color-encoded single-shot computer-generated Moiré profilometry}},
  booktitle = {Scientific},
  year = {2021},
  doi = {10.1038/s41598-021-90522-x},
}
Figure