Image Correction via Foreground-Background Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods struggle to effectively correct non-uniform illumination in images due to their inability to fully utilize foreground information and accurately reflect background illumination changes.
Innovation Solution
An image correction method and apparatus that identifies pixels as foreground or background, estimates background brightness based on adjacent pixel gradients and saturation, and generates a background illumination map to correct image brightness, using a combination of HSV color space conversion, gradient, and saturation analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional filtering or fitting methods are used to estimate background as a whole, then the processing is simple, but the foreground information cannot be utilized fully and it is difficult to reflect illumination change of the background properly
Solution Approach 1:
The image is segmented into foreground and background regions using HSV color space conversion and saturation/gradient analysis. This segmentation allows separate processing of foreground and background, enabling accurate background illumination estimation without being affected by foreground objects, thus resolving the contradiction between estimation accuracy and processing complexity
Solution Approach 2:
Instead of estimating background as a whole, the method estimates background illumination locally for each background pixel based on its adjacent pixels. This local quality approach allows the background estimation to adapt to local illumination changes while maintaining reasonable processing complexity through targeted computation
2Adaptability or versatility
If background is estimated as a whole by filtering or fitting, then the processing complexity is low, but the illumination change of the background cannot be reflected properly
Solution Approach 1:
The background illumination estimation is made dynamic by allowing each background pixel to be estimated based on its local neighborhood rather than using a fixed global model. This dynamic approach enables the system to adapt to varying illumination conditions across different regions of the image while maintaining computational efficiency through localized processing
Data Source
AI summary
An image correction method and an image correction apparatus when the image correction method includes: an identifying step of identifying each pixel in an image as a foreground pixel or a background pixel; a background filling step of estimating brightness of a background corresponding to a foreground pixel based on brightness and gradient of the brightness of background pixels adjacent to the foreground pixel to fill the background located in a position of the foreground pixel, to obtain a background illumination map of the image according to filled backgrounds along with background pixels; and a correcting step of correcting the image based on the brightness of each pixel in the image and the background illumination map. A non-uniform illumination image can be corrected effectively.