Quantization analysis of speckle intensity measurements for phase retrieval

Area of Science:

• Optical Metrology
• Digital Imaging
• Wavefront Sensing

Background:

• Phase retrieval (PR) relies on speckle intensity measurements.
• Digital cameras introduce quantization error, degrading signal quality.
• Quantization effects on PR are not fully understood.

Purpose of the Study:

• Investigate the impact of quantization error on speckle intensity distribution.
• Evaluate the influence of quantization on phase retrieval accuracy and convergence.
• Determine optimal bit depth for speckle-based PR in static wavefront sensing.

Main Methods:

• Numerical simulations of speckle intensity distributions.
• Experimental investigations using a static wavefront sensing setup.
• Analysis of phase retrieval performance across different bit depths.

Main Results:

• 3 to 4 bits are adequate for representing speckle intensities for acceptable PR.
• Acceptable reconstructions achieved with relatively fast convergence rates.
• Using 4-bit cameras reduces computer memory requirements by 2.4 times compared to 8-bit.

Conclusions:

• Lower bit depth (3-4 bits) is sufficient for speckle-based phase retrieval.
• Reduced memory footprint facilitates rapid speckle data acquisition.
• Findings support the use of lower-bit cameras for efficient dynamic wavefront sensing.