Image Blur Correction via Dynamic Filter Blending
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in performing effective blur correction, particularly due to lens aberration and varying blur patterns with image height, leading to inadequate correction and high memory requirements for storing multiple filters.
Innovation Solution
An image processing apparatus and method that calculates and uses a blend of default and coordinate position-corresponding high pass filters based on image height and focus position, reducing the number of filters needed and optimizing blur correction by adjusting blend rates and filter magnitudes accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple filters are stored in memory to correct different blur patterns according to image height, then correction accuracy is improved, but memory capacity requirement increases
Solution Approach 1:
The patent changes the parameters of a single default filter dynamically based on image height and focus position. Instead of storing multiple fixed filters, the system adjusts filter coefficients and characteristics in real-time according to the specific imaging conditions, thereby achieving accurate correction for different blur patterns while using minimal memory
Solution Approach 2:
The patent implements a dynamic filter generation mechanism where the blur correction filter is not fixed but adapts continuously based on image height and focus position. The system calculates appropriate filter parameters on-the-fly, transforming a static multiple-filter approach into a dynamic single-filter approach that provides the same correction accuracy with reduced memory requirements
2Quantity of substance
If a single default filter is used for all pixels, then memory requirements are reduced, but correction accuracy deteriorates due to varying blur patterns
Solution Approach 1:
The patent applies parameter changes to the default filter by adjusting its coefficients based on image height and focus position. This allows a single filter structure to adapt to different blur patterns, maintaining correction accuracy while using minimal memory
Solution Approach 2:
The patent makes a single default filter universal by enabling it to function for multiple different blur correction scenarios. Through dynamic parameter adjustment based on image height and focus position, one filter performs the work of what would traditionally require multiple specialized filters
3Measurement precision
If filters are corrected only by shifting center of gravity positions, then partial blur correction is improved, but control of correction magnitude is insufficient
Solution Approach 1:
The patent extends parameter adjustment beyond just center of gravity position to include multiple filter characteristics such as coefficient values, shape parameters, and frequency response characteristics. This comprehensive parameter control enables both accurate positioning and proper magnitude control for blur correction
Data Source
AI summary
There is provided an image processing apparatus including an image signal correction unit for performing a blur correction process on an image. The image signal correction unit includes a blur correction processing section configured to perform a blend process for a default filter and a coordinate position-corresponding high pass filter, which is generated according to an image height of a pixel to be subjected to blur correction, using a blend coefficient determined according to the image height of the pixel to be subjected to blur correction, and to perform a blur correction process using a filter generated through the blend process.