Related Experiment Video
Updated: Aug 16, 2025
11:34
High-resolution, High-speed, Three-dimensional Video Imaging with Digital Fringe Projection Techniques
Published on: December 3, 2013
15.7K
Deep learning-enabled anti-ambient light approach for fringe projection profilometry.
Optics Express
|December 23, 2022
Summary
This study introduces a deep learning-enabled anti-ambient light (DLAL) approach for fringe projection profilometry (FPP). DLAL effectively extracts 3D surface profiles even under strong ambient light by suppressing interference, enabling high-quality imaging.
Area of Science:
- Optical Metrology
- Computer Vision
- 3D Imaging
Background:
- Fringe projection profilometry (FPP) struggles with ambient light interference, reducing stripe quality and 3D reconstruction accuracy.
- Conventional methods like large-step phase shifting are time-consuming and less effective under strong illumination.
Purpose of the Study:
- To develop a novel deep learning-enabled anti-ambient light (DLAL) approach for robust FPP under challenging lighting conditions.
- To model and mitigate ambient light-induced phase error (ALPE) in FPP systems.
Main Methods:
- A deep learning model was trained on a custom dataset generated using an ambient light-induced phase error (ALPE) model.
- The DLAL approach extracts phase distributions from a single fringe image, suppressing outliers caused by ambient light.
Main Results:
- The DLAL approach effectively suppresses outliers and enables high-quality 3D surface imaging in the presence of strong ambient light.
- Experimental validation confirmed the effectiveness and adaptability of DLAL in both indoor and outdoor scenarios.
Conclusions:
- Deep learning offers a promising solution for overcoming ambient light limitations in FPP.
- The proposed DLAL method significantly enhances the performance and applicability of FPP for 3D surface profiling in real-world environments.
