Holographic Object Sorting via Direct Characteristic Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital holographic microscopy for particle characterization is limited by high computational costs due to the need for extensive numerical algorithms involving inversion, nonlinear pattern matching, and image analysis decomposition, which restricts real-time characterization in high-throughput applications.
Innovation Solution
A device and method utilizing a light sensor with intersecting rows of light recording elements forming a cross-shape, coupled with a processing unit that extracts object characteristics directly from a section of the hologram without reconstructing the image, employing machine learning components like artificial neural networks or support vector machines, and capturing intensity and phase information to reduce computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital holographic microscopy uses extensive numerical algorithms for image reconstruction, then measurement precision is improved, but computational cost increases significantly
Solution Approach 1:
The patent extracts only the essential information needed for object characterization directly from the hologram without performing complete image reconstruction. By using support vector machines to identify objects and extract characteristics from raw holographic data, the system eliminates the computationally expensive inversion and nonlinear pattern matching steps while maintaining characterization accuracy.
Solution Approach 2:
The patent replaces traditional mechanical image reconstruction algorithms with a machine learning-based approach using support vector machines. This substitution transforms the computational process from extensive numerical algorithms involving inversion to a more efficient classification and characteristic extraction process that achieves the same measurement precision with lower computational cost.
2Measurement precision
If traditional algorithms perform complete image reconstruction, then measurement precision is improved, but processing speed decreases
Solution Approach 1:
The patent extracts only the necessary characteristic information directly from the hologram data without performing complete image reconstruction. By using support vector machines to directly process holographic data and extract particle characteristics, the system eliminates time-consuming reconstruction steps while maintaining characterization precision, thereby significantly improving processing speed for high-throughput applications.
Solution Approach 2:
The patent performs preliminary classification and characteristic extraction using support vector machines before any detailed image reconstruction would be needed. This preliminary action identifies objects and extracts their characteristics directly from the hologram, enabling fast processing while maintaining the precision needed for subsequent analysis if required.
3Measurement precision
If extensive numerical algorithms are used for hologram analysis, then measurement precision is improved, but energy consumption increases
Solution Approach 1:
The patent extracts essential object characteristics directly from the hologram without performing complete image reconstruction through extensive numerical algorithms. By using support vector machines to directly process holographic data and extract particle properties, the system eliminates the energy-intensive computation steps while maintaining measurement precision, thereby significantly reducing power consumption for real-time characterization.
Solution Approach 2:
The patent replaces energy-consuming traditional numerical algorithms with a machine learning-based support vector machine approach. This substitution reduces computational complexity and energy consumption while maintaining the precision needed for accurate object characterization in high-throughput applications.
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 accurate characterization of objects at low computational cost, allowing for real-time classification of cells at high speeds, such as 100,000 to 1 million objects per second, with reduced power consumption and increased accuracy.
Implementation Method 1
In digital holographic microscopy, light wave front information from an illuminated object is digitally recorded as a hologram
Implementation Method 2
the light sensor consists of two rows of light recording elements intersecting each other thereby forming a cross-shape
Data Source
Figure 1~2
Figure 3A~3B
Figure 3C~3E
AI summary
A device for extracting at least one object characteristic of an object (106) is presented, the device comprising: a light sensor (101) for recording a hologram of an object and a processing unit (102) coupled to the light sensor and configured for extracting at least one object characteristic from the hologram; wherein the processing unit is configured for extracting the at least one object characteristic from a section of the hologram without reconstructing an image representation of the object. Further, a device (200) for sorting an object (106), a method for identifying an object and a method for sorting objects is presented.