Optical Scattering Imaging

Basic Principles of Optical Scattering Imaging:
When light encounters scatterers (e.g., turbid media, biological tissues) during propagation, its direction changes—a phenomenon known as scattering. In optical scattering imaging, photons carrying target information are disrupted by the inhomogeneous distribution of particles and refractive indices within the scattering medium, leading to distorted direct-detection images. For instance, in foggy conditions, light scattering by water droplets causes blurred observation of objects. However, optical scattering imaging relies on analyzing and processing these scattered photons to reconstruct images. Scattered photons can be categorized as:

• Ballistic photons (traveling nearly straight, carrying clear target information),

• Snake-like photons (undergoing multiple scattering, retaining partial target information),

• Diffuse photons (highly randomized after extensive scattering).
  Different photon types play distinct roles in imaging. Traditional scattering imaging techniques often focus on optimizing the collection of ballistic photons for image reconstruction.

Traditional Optical Scattering Imaging Techniques:
Conventional methods based on ballistic photon collection attempt to extract target information by isolating these photons from scattered light. Early approaches utilized specific optical designs and detector configurations to prioritize ballistic photon capture. However, in practical scenarios, ballistic photons are scarce, and most photons in strongly scattering media are non-ballistic due to multiple scattering. Consequently, such techniques perform poorly in media with large optical thickness and have limited applicability.

Computational Optical Scattering Imaging:
With technological advancements, computational scattering imaging has emerged, emphasizing the utilization of non-ballistic photons in thick scattering media. Key approaches include:

• Optical Memory Effect and Phase Retrieval Algorithms:
  The optical memory effect describes how scattering media retain "memory" of incident light under certain conditions—small changes in illumination angle or position produce correlated variations in the scattered field. Leveraging this effect with phase retrieval algorithms enables recovery of target phase information from scattered fields. For example, experiments reconstruct target images by correlating scattered light with targets through the memory effect and iteratively solving phase information. This method shows promise for dynamic thick scattering media and potential in wide-field, long-range imaging.

• Coherent Diffraction Imaging:
  This technique employs coherent light illumination and iterative algorithms to reconstruct target amplitude and phase from measured diffraction patterns. By recording scattered light intensity (lacking phase data), phase retrieval algorithms iteratively solve for missing information. Coherent diffraction imaging surpasses traditional resolution limits, enabling high-resolution imaging of microstructures in materials science and biomedicine.

• Ptychographic Iterative Engine:
  Ptychography reconstructs high-resolution images by overlapping scans of target regions and iteratively processing scattered intensity data. Continuously adjusting scan positions and angles enhances information acquisition, improving resolution and quality. This method excels in imaging non-sparse targets and holds significant value in practical scattering imaging applications.

Challenges and Limitations:
Despite notable progress, optical scattering imaging faces challenges:

• Dynamic environments: Rapidly changing scattering media (e.g., flowing smoke, dynamic biological tissues) demand real-time processing of evolving scattering data, requiring highly efficient algorithms and computational power.

• Resolution and quality: Thick scattering media often degrade image quality due to information loss and noise from multiple scattering, leading to blurring or distortion.

• Scenario specificity: Many techniques excel in specific conditions but lack generalizability, limiting their robustness across diverse real-world applications.

Applications:

• Biomedicine: Enables imaging of internal tissue structures through light scattering, aiding disease diagnosis (e.g., detecting early-stage cancer via analysis of scattered light from tissues).

• Environmental Monitoring: Facilitates imaging through fog, smoke, or haze to monitor distant pollution sources or meteorological phenomena.

• Industrial Inspection: Supports non-destructive testing of opaque materials by analyzing scattered light to identify internal defects, enhancing product quality and safety.