Three-dimensional single-pixel imaging with far fewer measurements than effective image pixels

Area of Science:

• Optics and Photonics
• Computational Imaging
• 3-D Reconstruction

Background:

• Traditional single-pixel imaging demands extensive measurements for high-resolution 3-D reconstruction.
• Achieving both high sampling efficiency and depth accuracy in 3-D single-pixel imaging remains a significant challenge.

Purpose of the Study:

• To develop a highly efficient and accurate 3-D single-pixel imaging technique.
• To overcome the limitations of conventional methods in sampling efficiency and depth accuracy.

Main Methods:

• Virtual fringe projection implemented via Helmholtz reciprocity.
• Depth information encoded into deformed fringe patterns.
• Sinusoidal intensity pattern illumination and single-pixel detection for Fourier spectrum sampling.
• Fourier analysis for 3-D information retrieval.

Main Results:

• Experimental 3-D reconstruction of a complex object with 599×599 effective pixels.
• Achieved a measurement-to-pixel ratio of 5.78%, demonstrating high sampling efficiency.
• Sub-millimetric depth accuracy validated using a test object.

Conclusions:

• The proposed method significantly enhances sampling efficiency in 3-D single-pixel imaging.
• It offers a practical solution for high-accuracy 3-D reconstruction with reduced data acquisition.
• This technique holds promise for applications requiring efficient and precise 3-D imaging.