Image Contrast Adjustment Using Region-Specific JND and Background Brightness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for adjusting image contrast, such as converting images to HSV or LAB color space and adjusting luminance curves, fail to fully preserve image details and are not universally applicable, leading to poor contrast adjustment effects.
Innovation Solution
A method that involves acquiring a grey-scale image, dividing it into regions using a watershed algorithm, calculating Just Noticeable Difference (JND) values, determining target contrast based on region brightness and JND values, and adjusting the image contrast accordingly, considering human perception of brightness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If contrast adjustment is performed by converting to HSV or LAB color space and adjusting luminance curves, then the overall contrast of the image can be adjusted, but the details of the image cannot be fully preserved and the distribution of light and dark areas is undermined
Solution Approach 1:
The image is divided into multiple local regions using watershed algorithm, allowing contrast adjustment to be performed separately in each region rather than globally. This segmentation enables preservation of local details while adjusting overall contrast, as each region's luminance curve can be optimized independently based on its characteristics.
Solution Approach 2:
Different luminance curves are applied to different local regions of the image based on their specific characteristics. Instead of using a single global luminance curve, the invention adjusts contrast locally in each divided region, preserving the unique details and light-dark distribution of each area while achieving overall contrast improvement.
2Adaptability or versatility
If a single Gamma parameter is used to adjust contrast, then the contrast adjustment can be applied to all images, but the setting cannot be optimized for specific images and details are lost
Solution Approach 1:
The invention transitions from static single Gamma parameter adjustment to dynamic region-specific luminance curve adjustment. Each local region can have its own optimized luminance curve selected from multiple candidate curves, enabling the system to adapt to different image characteristics while maintaining universal applicability across various image types.
Solution Approach 2:
Instead of using a fixed Gamma parameter, the invention employs multiple luminance curves with different characteristics. The system can select and apply appropriate luminance curves to different regions based on image content, achieving both universal applicability and precise contrast adjustment for specific images.
Data Source
AI summary
A method for adjusting contrast includes: acquiring a corresponding grey-scale image of a current image; dividing the grey-scale image to obtain a plurality of divided regions; calculating JND values corresponding to the divided regions based on an area of each of the divided regions; and calculating a target contrast based on the JND value of each of the divided regions and a background brightness of the current image; adjusting the contrast of the current image based on the target contrast and a predetermined condition.