Image Artifact Removal via Morphological Masking and Local Blurring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods for removing dust and impurities from camera lenses often degrade image quality and can cause smearing or incongruity between dust regions and their periphery, hindering effective correction of image artifacts.
Innovation Solution
An image processing method and apparatus that uses median window filters to obtain smoothed images, performs point-wise multiplication, generates a binary image through residual subtraction, applies morphological erosion, and uses a rim portion as a mask for blurring to correct images affected by dust and other impurities, thereby enhancing image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional dust correction methods are applied to remove impurities from images, then dust and impurities are removed from the image, but image quality degrades and smearing occurs in the object regions
Solution Approach 1:
The patent segments the image processing into multiple distinct stages: obtaining initial image data, generating a dust mask through morphological operations, identifying rim portions, and selectively applying blurring only to rim regions while preserving object regions. This segmentation allows different processing strategies to be applied to different parts of the image, removing dust while protecting object integrity.
Solution Approach 2:
The patent applies local quality by differentiating between object regions and rim portions, applying blurring operations only to rim portions where dust artifacts are most prominent. The processor selectively processes different regions with different operations, maintaining high quality in object regions while correcting artifacts in rim regions.
2Object-affected harmful factors
If conventional dust correction methods are applied to remove impurities, then dust regions are corrected, but a sense of incongruity is generated between the dust region and its periphery
Solution Approach 1:
The patent changes processing parameters based on location, applying blurring with specific window sizes (e.g., 51x51 or 9x9 pixels) only to rim portions identified through morphological erosion. This parameter change ensures that dust regions are corrected with appropriate smoothing while maintaining visual consistency with surrounding areas.
3Object-affected harmful factors
If simple blurring is applied to dust regions, then dust artifacts are reduced, but overall image quality and detail are lost
Solution Approach 1:
The patent applies local quality by restricting blurring operations exclusively to rim portions of the image, identified through morphological erosion operations. Object regions and central portions of the image remain unblurred, preserving fine details and information while still correcting dust artifacts in the rim regions where they are most problematic.
Data Source
AI summary
The present disclosure relates to method and apparatus for image recovery and image correction. In an embodiment, the present disclosure relates to recovering images that are occluded due to some material in the optical path while capturing the images. The detection of noisy region is performed by taking the residual image between the images yielded by large sized and small sized Gaussian window smoothening filters on red, green and blue channels. The recovery of the image that alleviates the noise is obtained by point-wise multiplication of the above said residue image with a suitably synthesized Gaussian distributed co-efficient of same size.