Ghost imaging in convolution space using shifted random-pattern illumination

Area of Science:

• Optics and Photonics
• Computational Imaging
• X-ray Imaging

Background:

• Ghost imaging enables image reconstruction using only single-pixel detectors.
• Conventional speckle-shifting ghost imaging requires multiple pattern shifts, limiting measurement efficiency.
• Accurate acquisition of convolution images is crucial for various imaging applications.

Purpose of the Study:

• To develop a more efficient ghost imaging technique for acquiring convolution images.
• To improve image quality and measurement efficiency over existing methods.
• To demonstrate the technique's potential in X-ray ghost imaging and invisible-light regimes.

Main Methods:

• Proposed a novel ghost imaging approach using shifted random-pattern illumination.
• Developed a theoretical framework to analyze the technique's performance.
• Conducted experiments to validate the method and compare it with speckle-shifting ghost imaging.

Main Results:

• The proposed method approximates ghost imaging in convolution space with improved efficiency.
• Theoretical analysis and experimental results show higher image quality compared to speckle-shifting ghost imaging.
• Experimental validation in X-ray ghost imaging was successful.

Conclusions:

• The shifted random-pattern illumination ghost imaging technique offers enhanced efficiency and image quality.
• This method overcomes the limitations of conventional speckle-shifting ghost imaging.
• The technique shows promise for high-speed imaging applications in invisible-light regimes, particularly in X-ray imaging.