Iterative Image Sharpness Correction for Full-Depth Focus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capture methods struggle to achieve sharpness without compromising on exposure time or requiring multiple images, especially when dealing with moving scenes or varying luminosity, leading to blurred foregrounds and backgrounds.
Innovation Solution
A method that corrects image sharpness by applying a function F(d0, Z0, a, b) or its inverse invF(d0, Z0, a, b) to each point or area of the input image, considering the imaging system's response, distance, and coordinates, to produce a rendered image with improved sharpness and resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the aperture diameter of the lens is reduced to increase depth of field and sharpness, then the image sharpness and depth of field are improved, but the light-gathering capability decreases and image capture time increases
Solution Approach 1:
The patent segments the image processing into multiple focus layers through focus bracketing, capturing images at different focal distances and combining them to achieve full sharpness without requiring a small aperture throughout the entire exposure
Solution Approach 2:
The system performs preliminary focus adjustment across multiple images before final combination, pre-establishing sharp focus at different depths to eliminate the need for long exposure times associated with small aperture shooting
2Manufacturing precision
If the aperture diameter of the lens is reduced to increase depth of field, then the sharpness at different depths is improved, but the luminosity of the scene decreases
Solution Approach 1:
The patent divides the scene into multiple focal planes and captures separate images for each plane using larger aperture settings, then combines them to achieve both high luminosity and full-depth sharpness
Solution Approach 2:
The system adds the depth dimension through multi-focus image bracketing, capturing images at different focal distances to achieve sharpness throughout the entire depth range while maintaining high aperture for better luminosity
3Productivity
If the aperture diameter of the lens is increased to improve luminosity and reduce capture time, then the light-gathering capability and speed are improved, but the depth of field and sharpness of foreground and background decrease
Solution Approach 1:
The patent segments the focal range into multiple discrete focus distances, capturing images at each distance with high aperture for speed, then combines them to achieve both fast capture and complete sharpness
Solution Approach 2:
The system performs preliminary focusing at multiple predetermined distances before combination, pre-capturing sharp images at different planes using fast high-aperture settings to maintain productivity
4Manufacturing precision
If multiple images are captured at different focus distances to achieve sharpness throughout the scene, then the depth of field coverage is improved, but the complexity of the imaging system increases
Solution Approach 1:
The patent uses a single imaging system that performs multiple functions: capturing images at different focal distances through bracketing and then combining them computationally, eliminating the need for multiple specialized lenses or complex mechanical focus-stacking mechanisms
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
The method achieves sharp images in all parts of the scene, even in low-light conditions and with moving subjects, without the need for long exposure times or multiple captures, maintaining a high numerical aperture and reducing image defects like aberrations and blur.
Implementation Method 1
a function F(d0, Z0, a, b) describing the response of the at least one imaging system... on the distance Z0 between, on the one hand, the imaged part corresponding to this point or area
Data Source
AI summary
A method for correcting at least one input image Ie into a rendered image IRk and then into a rendered image IR, each image originating from an optical sensor provided with photosites of different colors and being obtained through an optical imaging system, each sensor associated with an optical imaging system, the method involving:receiving each input image Ie,iteratively modifying the image IRk being rendered at different iterations k, by iteratively processing of a function E comprising two terms, i.e. a first term D, which depends on a comparison between each input image Ie and a result Ick of the image IRk being rendered at the iteration k reprocessed by information relating to the imaging system, and a second term P, which depends on anomalies or penalties or defects within the image IRk being rendered at the iteration k, until the function E is reduced.