Autofocus System Using Spatial Frequency Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autofocus techniques, such as contrast detection AF, face challenges in accurately determining the focus position when multiple contrast peaks are present, particularly when observing phase objects like cultured cells in a culture vessel, where varying parameters like vessel shape, solution depth, and interface shape complicate geometric calculations.
Innovation Solution
An observation system that illuminates the object from multiple directions and calculates the focus position based on image groups acquired at different optical axis positions, using cross-correlation functions and shift amount calculations to determine the focus position without requiring detailed parameter information in advance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If contrast detection AF is used to calculate focus position, then the method does not require dedicated sensors and is widely applicable, but it becomes difficult to accurately identify the focus position when multiple contrast peaks are formed in the optical axis direction
Solution Approach 1:
The patent transitions from one-dimensional contrast analysis along the optical axis to two-dimensional spatial frequency analysis. By computing power spectral densities from images taken at different lateral positions and analyzing the spatial frequency domain, the system can distinguish between different contrast peaks that appear similar in the spatial domain, thereby resolving the ambiguity in focus position identification while maintaining broad applicability
Solution Approach 2:
The patent introduces spatial frequency analysis as an intermediary between image acquisition and focus position determination. Instead of directly comparing contrast values at different depths, the system transforms images into the frequency domain, where the interference patterns created by multiple contrast peaks become distinguishable. This intermediary transformation enables accurate focus detection even when multiple peaks are present
2Measurement precision
If geometric autofocus calculation is used based on positional relationships between images, then focus position can be calculated without contrast detection, but detailed parameter information such as vessel shape, solution depth, and interface shape must be known in advance
Solution Approach 1:
The patent enables the system to automatically determine focus position using only lateral position information of the objective, without requiring external parameter inputs about the sample or culture vessel. The method extracts all necessary information from the acquired images themselves through spatial frequency analysis, making the system self-sufficient and eliminating the need for complex pre-characterization of the observation environment
Solution Approach 2:
The patent changes the fundamental parameter used for focus detection from contrast magnitude (which requires detailed sample parameters) to spatial frequency characteristics (which can be extracted directly from images). By transforming the problem from analyzing image intensity contrasts to analyzing spatial frequency patterns, the system achieves parameter-free operation while maintaining high precision
3Measurement precision
If images are acquired at multiple observation positions in the optical axis direction, then more information is available for focus calculation, but the time and number of images required increases
Solution Approach 1:
The patent extracts the essential information needed for focus determination from just two images taken at different lateral positions, rather than requiring multiple images at different optical axis positions. By taking out and analyzing the spatial frequency characteristics from this minimal set of images, the system achieves accurate focus detection with reduced imaging time and processing overhead
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 allows for accurate focus position calculation even in environments with multiple contrast peaks, enabling precise imaging of samples with varying heights and shapes without pre-defined parameter settings.
Implementation Method 1
an imaging device that includes an objective that condenses light from the observation object and images the observation object with the light collected by the objective
Implementation Method 2
an illumination device that illuminates an observation object with illumination light from a plurality of different directions
Data Source
AI summary
An observation system includes: an illumination device that illuminates an observation object with illumination light from a plurality of different directions; an imaging device that includes an objective that condenses light from the observation object and images the observation object with the light collected by the objective; and a control unit. The control unit calculates a focus position of the imaging device based on a plurality of image groups acquired by the imaging device at observation positions different from each other in an optical axis direction of the objective, each of the plurality of image groups including a plurality of images of the observation object illuminated from directions different from each other by the illumination device.