Regularized inverse holographic volume reconstruction for 3D particle tracking

Area of Science:

• Optics and Photonics
• Computational Imaging
• Fluid Dynamics

Background:

• Digital inline holography (DIH) faces challenges in particle tracking, including limited longitudinal resolution, particle concentration constraints, and complex processing.
• Current DIH methods struggle with accurate three-dimensional particle field reconstruction.

Purpose of the Study:

• To develop an advanced inverse problem method for full volumetric reconstruction in DIH.
• To address the limitations of poor longitudinal resolution and particle concentration limits in DIH.
• To reduce the computational cost of DIH reconstruction for practical applications.

Main Methods:

• Utilized an inverse problem approach with fused lasso regularization for particle field reconstruction.
• Exploited data sparsity and employed GPU processing to accelerate computational efficiency.
• Validated the method using both synthetic and experimental holographic data.

Main Results:

• Achieved accurate full volumetric reconstructions of particle fields.
• Demonstrated significant reduction in computational cost compared to traditional inverse methods.
• Successfully applied the method to high-concentration microorganism swimming and microfiber rotation.

Conclusions:

• The proposed inverse problem method significantly enhances DIH capabilities for particle tracking.
• This approach overcomes key limitations, enabling new quantitative analyses in complex scenarios.
• The method extends the applicability of DIH to previously intractable research areas.