Multiple moving object single-pixel imaging with a calibrated radon spectrum.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Signal Processing

Background:

• Single-pixel imaging (SPI) is limited by motion blur in dynamic scenes.
• Existing SPI methods for moving objects typically address only a single object.
• High-speed motion in complex scenarios poses a significant challenge for SPI applications.

Purpose of the Study:

• To develop an SPI method capable of imaging multiple objects with distinct motion trajectories.
• To overcome the motion blur limitation inherent in SPI for dynamic scenes.
• To enable blur-free reconstruction of coexisting moving objects in complex environments.

Main Methods:

• Acquisition of a mixed radon spectrum from multiple moving objects using SPI.
• Implementation of a post-processing pipeline involving spectrum separation and calibration.
• Application of reconstruction algorithms to obtain individual object images.
• Evaluation across diverse scenarios with varying object types, trajectories, and velocities.

Main Results:

• Successful acquisition of mixed radon spectra for scenes with two distinct moving objects.
• Effective separation and calibration of individual object spectra.
• Generation of blur-free reconstructed images for each moving object.
• Demonstrated feasibility and robustness across various test conditions.

Conclusions:

• The proposed SPI method successfully addresses the challenge of imaging multiple moving objects simultaneously.
• The technique provides blur-free images, enhancing the applicability of SPI in dynamic environments.
• This approach offers a robust solution for scenarios involving complex object motions and trajectories.