All-in-focus Image Generation via Depth-Resolved Sharpness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods fail to generate vivid all-in-focus images when multiple objects are distributed in a complicated manner, as they rely on a single image reference value per pixel, leading to missing or blurred overlapping parts of cells in microscopic imaging.
Innovation Solution
An image processing method that calculates individual sharpness values at each focus position, determines corrective sharpness and image reference values, and combines these to generate an all-in-focus image by calculating luminance values based on multiple image reference values and corrective sharpness for each pixel, effectively addressing the issue of overlapping cells.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single image reference value is used for each pixel, then the processing is simple and fast, but overlapping cells are missed or blurred in the all-in-focus image
Solution Approach 1:
The patent divides the single image reference value into multiple image reference values corresponding to different depth positions. For each pixel, it identifies multiple candidate cells at different depths and assigns separate image reference values to each, enabling accurate representation of overlapping cells while maintaining processing efficiency through structured segmentation of the reference value assignment process
Solution Approach 2:
The patent introduces a depth dimension to the image reference value assignment. Instead of using a single reference value per pixel, it assigns multiple reference values corresponding to different focal planes, transforming the two-dimensional pixel mapping into a three-dimensional space that accounts for depth information and resolves overlapping cell issues
2Measurement precision
If multiple image reference values are assigned to each pixel, then overlapping cells are accurately represented, but the processing complexity increases
Solution Approach 1:
The patent applies local quality by assigning different image reference values to different regions based on their depth positions. For each pixel, it determines which cells are locally significant (overlapping cells at different depths) and assigns appropriate reference values only to those, avoiding unnecessary complexity in regions without overlapping structures
Solution Approach 2:
The patent changes the parameter of image reference value from a single fixed value to multiple variable values that adapt to the depth position of detected cells. This parameter change enables the system to handle varying depth configurations dynamically while maintaining a clear and manageable processing framework
Data Source
AI summary
Sharpness at each focus position is calculated as an individual sharpness. Thereafter, one or M (where M is a natural number equal to or greater than 2 and smaller than N) corrective sharpness are calculated and image reference values are determined from the individual sharpness. An luminance value is calculated based on the image reference values and the corrective sharpness for each pixel. An all-in-focus image is generated by combining those luminance values.