Lens Shading Color Correction via Block Matching Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lens shading color correction methods in digital cameras, particularly in miniaturized devices like smartphones, are inadequate due to inaccurate light attenuation correction, especially in images with complex textures and varying brightness, leading to visible distortions and color artifacts.
Innovation Solution
The method employs dynamic block matching and clustering to identify smooth areas in images, using self-similarity analysis and motion vectors to form clusters for accurate lens shading color correction, incorporating both global and local statistics analysis to improve data utilization and robustness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional lens shading color correction methods are used, then the correction process is simple, but the accuracy of light attenuation correction is insufficient leading to visible distortions and color artifacts
Solution Approach 1:
The patent segments the image into multiple blocks and performs block matching to identify corresponding regions. This segmentation approach enables accurate measurement of light attenuation by comparing pixel values across matched blocks, thereby improving correction precision while maintaining manageable computational complexity through localized processing.
Solution Approach 2:
The patent performs preliminary block matching and cluster formation before executing the actual lens shading correction. By pre-identifying smooth areas and establishing block correspondences, the method prepares accurate reference data that guides the subsequent correction process, ensuring high precision without requiring complex real-time calculations during correction.
2Volume of moving object
If miniaturization is pursued to reduce form factor, then user convenience increases, but lens and filter positioning causes color artifacts and discolorations
Solution Approach 1:
The patent replaces mechanical/optical correction approaches with a computational image processing solution. Instead of physically adjusting lens or filter positions to eliminate artifacts, the method uses block matching and cluster-based analysis to digitally correct color distortions, thereby maintaining the compact form factor while eliminating harmful color artifacts through software-based correction.
3Measurement precision
If block matching and clustering are used to identify smooth areas, then correction accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies local quality analysis by focusing block matching and cluster formation specifically on smooth areas of the image rather than processing the entire image uniformly. This localized approach improves distortion measurement accuracy in critical regions while reducing overall computational complexity by excluding complex texture areas from intensive processing.
Solution Approach 2:
The patent performs block matching and clustering operations selectively on portions of the image that contain smooth areas, rather than applying the full computational process to every region. This partial action approach achieves sufficient correction accuracy in relevant areas while avoiding unnecessary computational expenditure in regions where such processing would not improve results.
Data Source
AI summary
Systems, articles and methods to provide lens shading color correction using block matching are disclosed. Example processor systems disclosed herein are to process at least one cluster of blocks of an image to determine at least one modification parameter, modify the first shade correction data based on the at least one modification parameter to determine second shade correction data, and correct a lens shade effect associated with the image based on the second shade correction data.