3D Super-Oscillatory Diffractive Neural Network for Super-Resolution Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical super-oscillation methods face limitations such as small fields of view due to strong side lobes, short working distances, limited depth-of-focus, and chromatic aberration, which hinder practical applications in super-resolution microscopic imaging.
Innovation Solution
A super-oscillatory diffractive neural network (SODNN) is trained to optimize diffractive units in a three-dimensional space, generating a super-oscillatory focal spot with a large field of view, zero side lobes, and a long depth of field, and achromatism, thereby achieving high-performance super-resolution imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional optical super-oscillation methods are used, then super-resolution imaging is achieved, but the field of view is limited due to strong side lobes
Solution Approach 1:
The patent divides the optical field into multiple independent diffractive units arranged in a grid pattern, where each unit can be independently optimized. This segmentation allows the system to achieve super-resolution through collective optimization while controlling side lobes in each local region, thereby expanding the effective field of view.
Solution Approach 2:
The patent applies different optimization strategies to different regions of the optical field. By optimizing each diffractive unit locally to minimize side lobes in its specific region while maintaining super-resolution capability, the system achieves both high resolution and expanded field of view without the strong side lobes that limit conventional methods.
2Measurement precision
If conventional super-oscillation methods are used, then imaging is achieved, but the working distance is short
Solution Approach 1:
The patent transitions from conventional two-dimensional optical field optimization to three-dimensional optimization by introducing a depth dimension. The diffractive units are optimized to generate super-oscillatory fields that maintain resolution over an extended axial range, effectively achieving long working distance while preserving super-resolution capability.
3Measurement precision
If conventional methods are used, then imaging is achieved, but the depth of focus is limited
Solution Approach 1:
The patent performs preliminary optimization of the diffractive units to pre-establish a three-dimensional super-oscillatory field distribution that maintains resolution over an extended depth range. This preliminary design enables the system to achieve both high resolution and long depth of field without requiring post-processing adjustments.
4Measurement precision
If conventional optical super-oscillation is used, then imaging is achieved, but chromatic aberration occurs due to wavelength-dependent phase delay
Solution Approach 1:
The patent changes the optimization parameters from conventional wavelength-dependent phase control to a three-dimensional spatial frequency optimization approach. By optimizing the diffractive units in the spatial frequency domain rather than relying on wavelength-specific phase delays, the system achieves super-resolution while minimizing chromatic aberration across different wavelengths.
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
The SODNN effectively addresses the limitations of conventional methods by generating a super-oscillatory focal spot with a large field of view, zero side lobes, and a long depth of field, and reduces chromatic aberration, thereby enhancing the imaging capabilities.
Implementation Method 1
a super-oscillatory diffractive neural network (SODNN) is trained to optimize diffractive units in a three-dimensional space, generating a super-oscillatory focal spot
Implementation Method 2
Optical super-oscillation refers to the rapid sub-wavelength spatial variations of light intensity and phase that occur in complex electromagnetic fields formed by the precise interference of coherent light
Data Source
AI summary
A method and device achieving super-resolution microscopic imaging by a super-oscillatory diffractive neural network. By acquiring three-dimensional optical field constraint conditions, training a super-oscillatory diffractive neural network based on the three-dimensional optical field constraint conditions to optimize step heights of diffractive units in the super-oscillatory diffractive neural network, to minimize a difference of a light intensity distribution of a super-oscillatory focal spot and a light intensity distribution of side lobes generated by the super-oscillatory diffractive neural network from a light intensity distribution of an ideal output optical field, and/or to minimize light intensity outside a super-oscillatory region, and modulating incident light based on the trained super-oscillatory diffractive neural network to generate a super-oscillation effect in a three-dimensional space to acquire a super-resolution microscopic imaging result.