Frequency-Domain Image Correction for Turbulence and Motion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image correction methods, such as speckle imaging, require significant computational effort or specialized hardware, and struggle to correct image disturbances like air streaks and unintentional movements in real-time terrestrial imaging, especially when distinctive patterns are absent.
Innovation Solution
A method involving image processing that divides images into tiles, transforms them into spatial frequency space, identifies shifts by evaluating the spectral density function, compensates for these shifts, and reassembles the tiles, allowing for robust motion compensation with sub-pixel accuracy and reduced computational effort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If speckle imaging is used for image correction, then image restoration quality is improved, but computational time and hardware complexity increase significantly
Solution Approach 1:
The patent divides the image into multiple tiles or patches, processing each tile independently through the speckle imaging algorithm. This segmentation reduces the computational burden on each processing unit while maintaining overall image restoration quality, allowing parallel processing to decrease total computational time.
Solution Approach 2:
The patent applies speckle imaging correction selectively to regions of the image where atmospheric turbulence effects are most prominent, rather than uniformly processing the entire image. This partial action approach reduces unnecessary computational effort in areas where correction is less critical, decreasing overall processing time while maintaining restoration quality in affected regions.
2Measurement precision
If distinctive pattern registration is used for motion correction, then correction accuracy is improved when patterns are present, but the method fails when distinctive patterns are absent
Solution Approach 1:
The patent replaces the mechanical pattern-matching approach with a frequency-domain based speckle imaging method. Instead of relying on distinctive visual patterns for motion estimation, the method uses spectral analysis of image tiles to detect and correct atmospheric turbulence effects, making it effective regardless of whether distinctive patterns are present in the image content.
3Productivity
If conventional image processing is used, then processing speed is maintained, but image disturbances from atmospheric turbulence cannot be effectively corrected
Solution Approach 1:
The patent transforms the image data from the spatial domain to the frequency domain using Fourier transforms, enabling the detection and correction of atmospheric turbulence effects that are not visible in the spatial domain. This dimensionality change allows conventional processing speeds to be maintained while adding the capability to correct turbulence-induced image distortions through spectral analysis and filtering.
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
Enables efficient removal of unwanted motion from image sequences with sub-pixel accuracy, even in the absence of distinctive patterns, while preserving moving objects and reducing computational complexity.
Implementation Method 1
transforming, in particular by means of a Fourier transform, the tiles into a spatial frequency space
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The invention relates to methods, to devices, and to computer programs for image processing. In particular, a series of images is processed. The images are divided into tiles, and the tiles are transformed into the frequency domain. Disturbances, which are caused, for example, by air striations (shimmer), can be identified and rectified by evaluating the argument of the spectral density in the frequency domain.