Optical Imaging Using Spatial Modes Beyond the Diffraction Limit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing super-resolution imaging techniques are limited by the diffraction limit, requiring direct interaction with the sample or nonlinear properties, and are costly, making them unsuitable for certain applications like astronomical imaging or imaging of sensitive samples.
Innovation Solution
A method involving spatial mode sorting or heterodyne detection is used to process the spatial correlations of the optical field, leveraging higher-order spatial modes to achieve super-resolution without altering the sample, utilizing a neural network trained with theoretically expected and measured signals to reconstruct images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing super-resolution techniques (non-linear excitation, near-field probing) are used to achieve super-resolution, then resolution greater than the diffraction limit is achieved, but direct interaction with the sample is required and certain nonlinear properties of the sample are needed
Solution Approach 1:
The patent uses spatial correlations of the optical field as an intermediary to transfer sub-wavelength information from the object to the detector. Instead of directly interacting with the sample, the method processes the electromagnetic field's spatial correlations through coherent detection, allowing super-resolution without sample manipulation or nonlinear properties
Solution Approach 2:
The patent replaces mechanical or chemical interaction with samples (non-linear excitation, near-field probing) with optical field processing. By using spatial mode sorting and coherent detection of electromagnetic field correlations, the system achieves super-resolution through field manipulation rather than sample manipulation
2Measurement precision
If existing super-resolution systems are implemented, then super-resolution imaging is achieved, but the systems are costly to implement
Solution Approach 1:
The patent uses standard illumination and linear optics that are already available in conventional imaging systems. The spatial correlations contain the super-resolution information inherently, requiring only standard detection and processing, eliminating the need for expensive specialized equipment
Solution Approach 2:
The patent changes the detection parameter from direct intensity measurement to spatial correlation measurement. By detecting the electromagnetic field's spatial correlations through coherent detection and spatial mode sorting, the system extracts sub-wavelength information using standard optical components
3Ease of operation
If conventional direct intensity measurements are used, then the imaging process is simple, but the spatial correlations of the optical field are ignored and additional information about the object is lost
Solution Approach 1:
The patent extracts spatial correlation information from the optical field that would otherwise be lost in conventional intensity measurements. By using coherent detection and spatial mode sorting, the method separates and measures the correlation information contained in different spatial modes
Solution Approach 2:
The patent exploits the asymmetric information content in spatial correlations versus intensity. While intensity measurements provide symmetric averaging, spatial correlation measurements preserve the asymmetric positional information that encodes sub-wavelength object details
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
This approach achieves super-resolution imaging passively, reducing costs and enabling universal application across various scenarios, including those inaccessible to existing methods, by leveraging the fine spatial structure of spatial modes to access sub-wavelength information.
Implementation Method 1
the light field from an object being imaged experiences diffraction as it propagates through the optical components of the imaging system
Implementation Method 2
Alternatively, heterodyne detection using a local oscillator in the different modes can be used
Data Source
Figure 1
Figure 2A~2F
Figure 3
AI summary
A method (100) of training an image processing unit (27) for use in optical imaging, the method (100) comprising: providing (102) a training object (37); computing (104) a plurality of theoretically expected signals generated by detecting the component of the electromagnetic field arriving from the object (37) in each of a plurality of different spatial modes of light; and generating (106) a reconstructed image (39) based on the theoretically expected signals, wherein the reconstructed image (39) is provided as a label for the training object (37) for use in training the image processing unit (27).