Hologram Transfer Function Optimization for Optical Aberration Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise reduction algorithms fail to effectively reduce noise in incoherent holograms, leading to significant degradation of reconstruction quality due to increased bias signals and noise components from optical aberrations, especially with multiple point light sources.
Innovation Solution
A method involving constructing a dataset with ground truth images, optimizing a transfer function using gradient descent, and restoring noisy holograms with an optimized transfer function to minimize noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If incoherent holography is used to enable imaging under various light conditions, then adaptability is improved, but noise from optical aberrations increases
Solution Approach 1:
The patent applies preliminary action by pre-optimizing the transfer function using gradient descent before hologram restoration. The transfer function is trained on a dataset of hologram-image pairs to learn the optimal mapping that compensates for optical aberrations, enabling noise reduction to occur automatically during the restoration process without requiring real-time adjustments
Solution Approach 2:
The patent changes parameters by optimizing the transfer function parameters through gradient descent optimization. The transfer function parameters are adjusted to minimize the difference between restored holograms and ground truth images, thereby adapting the restoration process to compensate for specific optical aberrations in the system
2Manufacturing precision
If the number of point light sources increases to improve image detail, then manufacturing precision is improved, but noise components overlap and degradation increases
Solution Approach 1:
The patent converts the harmful effect of noise overlap into a benefit by using the optimized transfer function to learn and compensate for the specific noise patterns generated by multiple point light sources. The gradient descent optimization process enables the system to adapt to the noise characteristics and recover high-quality images even with significant noise overlap
3Device complexity
If conventional noise reduction algorithms are used to reduce computational complexity, then device complexity is reduced, but noise removal effectiveness is insufficient
Solution Approach 1:
The patent applies preliminary action by pre-optimizing the transfer function offline using gradient descent on training data. This preliminary optimization creates a robust restoration model that can be applied directly during actual hologram processing without requiring complex real-time optimization, achieving both simplicity and effectiveness
4Reliability
If deep-learning techniques are used to reduce hologram noise, then noise reduction effectiveness is improved, but inference speed becomes slow
Solution Approach 1:
The patent applies preliminary action by performing the complex optimization process offline before actual use. The transfer function is trained in advance on a dataset, and the optimized parameters are stored for rapid application during restoration, eliminating the need for slow real-time deep learning inference while maintaining high effectiveness
Solution Approach 2:
The patent uses a simplified transfer function model that, while less complex than full deep learning networks, provides sufficient performance for the specific task of hologram restoration. This lighter model enables faster processing while achieving comparable noise reduction effectiveness to more complex deep learning approaches
Data Source
AI summary
Disclosed herein is an apparatus and method for optimizing image quality of an acquired hologram. The method may include constructing a dataset in which a ground truth image is paired with hologram information acquired from the ground truth image, optimizing a transfer function based on a gradient descent method using the constructed dataset, and restoring noisy hologram information based on the optimized transfer function.