Digital Holographic Reconstruction Using Iterative Regularization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital holography methods face challenges in extracting depth information from samples due to out-of-focus optical signals and lack of phase information, leading to artifacts and shape distortion, especially when imaging fast-changing processes, and require a large number of holograms, limiting acquisition speed.
Innovation Solution
A method for robust digital holographic reconstruction using a limited number of interference patterns, enabling label-free three-dimensional imaging without fluorescent markers, employing a regularization term and forward-backward splitting method to handle non-differentiable terms, allowing for fast and high-quality imaging with a large field of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of holograms are acquired to enable phase retrieval and three-dimensional reconstruction, then imaging quality and accuracy are improved, but acquisition time increases significantly
Solution Approach 1:
The patent applies partial action by acquiring only a limited number of holograms (e.g., 4-8 angles) rather than comprehensive coverage. This partial sampling is sufficient for high-quality reconstruction when combined with the iterative algorithm and regularization term, thereby reducing acquisition time while maintaining imaging quality.
Solution Approach 2:
The patent implements feedback through an iterative algorithm that repeatedly adjusts the three-dimensional scattering potential based on the acquired holograms. The algorithm uses a regularization term to guide the iteration toward an accurate solution, enabling high-quality reconstruction from limited angular data through iterative refinement rather than requiring extensive initial data.
2Measurement precision
If fluorescent markers are used to enable three-dimensional imaging, then depth information extraction is improved, but sample health and viability are compromised
Solution Approach 1:
The patent extracts depth information directly from the phase components of light scattered by the sample itself, without requiring external fluorescent markers. By analyzing the phase information in the holograms and using iterative reconstruction algorithms, the system obtains three-dimensional scattering potential maps that reveal depth structure while leaving the sample unmodified and healthy.
Solution Approach 2:
The patent uses the phase information of scattered light as an intermediary carrier of depth information. Instead of relying on fluorescent markers to encode depth, the method extracts depth cues from the phase variations in the light waves themselves, which are naturally present in the scattering pattern and require no additional substances.
3Productivity
If a limited number of interference patterns are used for digital holographic reconstruction, then acquisition speed is improved, but imaging quality and accuracy deteriorate
Solution Approach 1:
The patent replaces the mechanical approach of acquiring many holograms through extensive angular scanning with a computational approach. An iterative algorithm processes the limited hologram data mathematically, using a regularization term to reconstruct high-quality three-dimensional images. This substitution of mechanical data collection with computational reconstruction enables fast acquisition while maintaining quality.
Solution Approach 2:
The patent changes the approach from acquiring many holograms at different angles to acquiring few holograms and compensating through parameter optimization in the reconstruction algorithm. The iterative algorithm adjusts the three-dimensional scattering potential parameters to best match the limited hologram data, achieving high-quality reconstruction from minimal angular samples.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables fast and high-quality three-dimensional imaging with a short acquisition time, suitable for quickly changing processes, and reduces computational resources, allowing for efficient imaging of biological samples with minimal distortion.
Implementation Method 1
a light beam providing uniform illumination of an object is used for creating an interference pattern based on object light, being scattered by the object, and reference light of the light beam passing unaffected through the object
Implementation Method 2
Digital holography uses digital image sensors, such as a charge-coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) image sensor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for three-dimensional imaging of a sample (302) comprises: receiving (102) interference patterns (208) acquired using light-detecting elements (212), wherein each interference pattern (208) is formed by scattered light from the sample (302) and non-scattered light from a light source (206; 306), wherein the interference patterns (208) are acquired using different angles between the sample (302) and the light source (206; 306); performing digital holographic reconstruction by applying an iterative algorithm to change a three-dimensional scattering potential, i.e. amplitude, of the sample (302) to improve a difference between the received interference patterns (208) and predicted interference patterns based on the three-dimensional scattering potential; wherein the iterative algorithm reduces a sum of a data fidelity term and a non-differentiable regularization term and wherein the iterative algorithm includes a forward-backward splitting method alternating between forward gradient descent (108) on the data fidelity term and backward gradient descent (110) on the regularization term.