Focusing Position Detection Using Image Segmentation and Saturation Exclusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional focusing position detection technologies face challenges in accurately detecting the focusing position, especially when object images are relatively dark or include high-luminance regions, leading to instability and potential errors due to background noise and the presence of dead cells.
Innovation Solution
A method involving the acquisition of object images while changing the focal position, with focus degrees calculated for local regions and saturation regions excluded to improve detection accuracy, allowing for stable focusing position detection without repeated imaging operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If imaging is performed with adjusted settings to increase luminance difference and suppress high-luminance regions, then detection accuracy improves, but phototoxicity and fluorescence photobleaching occur due to repeated imaging
Solution Approach 1:
The patent extracts and removes high-luminance regions (saturation regions) from the image before calculating focus degree. This eliminates the harmful influence of dead cells and other high-luminance objects on the focusing position detection, allowing accurate detection without needing to repeat imaging with adjusted settings that would cause phototoxicity and fluorescence photobleaching.
Solution Approach 2:
The patent divides the image into multiple local regions and calculates focus degree for each region separately. By segmenting the image, the system can identify and exclude high-luminance regions while still utilizing information from other regions, improving detection accuracy without requiring repeated imaging operations that would harm the specimen.
2Measurement precision
If conventional focusing position detection is used on dark fluorescent images, then detection speed is maintained, but accuracy deteriorates due to small luminance difference between object and noise
Solution Approach 1:
The patent divides the dark fluorescent image into multiple local regions and calculates focus degree for each region. This segmentation allows the system to accumulate information from multiple regions, improving the signal-to-noise ratio and detection reliability without requiring increased illumination that would cause photodamage.
Solution Approach 2:
The patent extracts and removes high-luminance regions that could skew the focus degree calculation. By excluding these regions and calculating focus degree only from appropriate local regions, the system achieves stable and accurate focusing position detection even in dark fluorescent images with small luminance differences.
3Measurement precision
If high-luminance regions are excluded to remove dead cell effects, then detection accuracy improves, but device complexity increases
Solution Approach 1:
The patent divides the image into multiple local regions, which simplifies the process of identifying and excluding high-luminance regions. By working with smaller local regions rather than the entire image, the computational complexity is reduced while still achieving accurate focusing position detection through systematic processing of segmented regions.
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 enables stable detection of the focusing position even in low-luminance or high-luminance scenarios, reducing the impact of background noise and dead cells, thereby improving accuracy and avoiding issues like phototoxicity and fluorescence photobleaching.
Implementation Method 1
acquiring a plurality of object images by imaging an imaging object by using an imager while changing a focal position along an optical axis
Implementation Method 2
the imaging object is irradiated with excitation light to emit fluorescence and then a fluorescent image (object image) including the imaging object image is acquired
Data Source
AI summary
In a focusing position detection method, a second step is carried out. The second step includes dividing the object image into a plurality of local regions, obtaining a local value indicating the degree of focalization from the local region for each of the plurality of local regions and obtaining the focus degree on the basis of the plurality of local values. Thus, even when a plurality of object images acquired by imaging an imaging object by using an imager while changing a focal position along an optical axis are relatively dark or include a high-luminance region, it is possible to stably detect a focusing position without repeating detection operations.