Image Refocusing via Gradient Domain PSF Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image refocusing techniques either result in slow processing or compromise image quality, particularly when dealing with high-resolution images or large depth ranges, as they either require multiple updates per pixel for high-quality images or use accelerating methods that sacrifice image quality for speed.
Innovation Solution
The technique involves converting the point spread function (PSF) to a gradient domain through differentiation, reducing nonzero elements to create a sparser PSF, which is then applied to each pixel to refocus the image in real-time without significant quality loss, using multi-order differentiation to achieve a maximum reduction in calculation and improve processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct computing method is used for image refocusing, then image quality is maintained, but processing speed is slow
Solution Approach 1:
The patent segments the continuous PSF into multiple discrete depth layers. Each layer is processed separately with its own PSF, allowing parallel computation and reducing the overall computational burden while maintaining image quality across different depth planes
Solution Approach 2:
The patent transforms the PSF from spatial domain to frequency domain using Fourier transform. This parameter change enables more efficient computation through frequency-domain convolution, significantly accelerating processing while preserving refocusing quality
2Productivity
If accelerating methods are used for image refocusing, then processing speed is improved, but image quality is compromised
Solution Approach 1:
The patent pre-computes and stores PSFs for multiple depth layers before actual refocusing is needed. These pre-computed PSFs are cached and reused during runtime, eliminating the need for repeated heavy computations while maintaining accurate refocusing quality
Solution Approach 2:
The patent implements adaptive PSF selection based on the actual depth of the target object. The system dynamically adjusts which pre-computed PSF to apply, optimizing both processing speed and image quality by using the most appropriate PSF for each specific refocusing scenario
3Measurement precision
If multiple updates per pixel are performed for high-quality images, then image quality is improved, but processing time increases
Solution Approach 1:
The patent applies partial updates to pixels based on their depth information. Only pixels corresponding to objects at the target depth plane undergo full refocusing updates, while other pixels use simpler transformation methods, reducing total processing time while maintaining quality for the region of interest
Data Source
AI summary
A system and method for refocusing an image including determining a point spread function (PSF) according to region of interest (ROI) and pixel depth, and converting the PSF to a gradient domain including differentiating the PSF to reduce nonzero elements in the PSF. The technique spreads intensity of pixels into a circle of confusion per the differentiated PSF. A shape of an optical system or aperture of the imaging device may be considered.