Single-pixel three-dimensional imaging based on 2D compressed sensing algorithm

Area of Science:

• Optics and Photonics
• Computer Vision
• Signal Processing

Background:

• Non-scanning 3D imaging is crucial for rapid detection and recognition.
• Compressed sensing (CS) enables signal reconstruction from minimal data, ideal for fast 3D imaging.

Purpose of the Study:

• To achieve single-pixel, non-scanning 3D imaging using a novel CS algorithm and time-of-flight.
• To evaluate the performance of the V2DALM algorithm in 3D imaging applications.

Main Methods:

• Developed the V2DALM algorithm, a new 2D compressed sensing approach.
• Integrated V2DALM with the time-of-flight measurement principle for 3D imaging.
• Verified feasibility through principle experiments and simulations.

Main Results:

• Algorithm V2DALM demonstrated significantly shorter reconstruction times compared to TVAL3 at similar imaging quality.
• V2DALM showed superior noise resistance over 2DPG-ED at the same sampling rate.
• Experimental results confirmed the feasibility of V2DALM for 3D imaging.

Conclusions:

• The V2DALM algorithm facilitates rapid, high-resolution depth map reconstruction for 3D objects.
• This non-scanning 3D imaging technique shows potential for applications in rapid detection and recognition.